Pith. sign in

REVIEW 4 major objections 3 minor 3 references

The Semigeostrophic--Euler Limit via Perturbative Monge--Amp\`ere Estimates

T0 review · 4 major / 3 minor · reviewed 2026-08-03 · deepseek-v4-flash

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

Pith's one-line read Small-amplitude semigeostrophic flow is shown to stay an O(ε) approximation of two-dimensional Euler in both velocity and density for a physical time of order ε^{-1} log log(1/ε).

desk verdict Solid Wasserstein comparison and a sound flow-based strategy, but the lifespan and O(ε) velocity theorems hinge on an unproved C^α estimate in Prop 5.3. read the letter →

arxiv 2601.04797 v3 pith:OTVKUGBP submitted 2026-01-08 math.AP physics.flu-dyn

Victor Armegioiu This is my paper
classification math.APphysics.flu-dyn MSC 35Q3135Q3535J9649Q22
keywords semigeostrophicequationsEulerMonge-Ampèresmall-amplitudelimitlifespanWassersteindistanceflowcomparisonoptimaltransport
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 aims to prove that, on the flat torus and in the small-amplitude regime, the semigeostrophic (SG) equations are a strong quantitative approximation of the two-dimensional incompressible Euler equations. It establishes a lifespan lower bound of order $\varepsilon^{-1} \log \log(1/\varepsilon)$, an $O(\varepsilon)$ bound in $L^2$ for the velocity difference on that window, and an $O(\varepsilon)$ Wasserstein bound for the associated physical densities. The engine is the expansion of the Monge-Ampère relation $\det(I+\varepsilon D^2 \psi) = 1 + \varepsilon \Delta \psi + \varepsilon^2 \det D^2 \psi$, which turns SG into a Poisson equation with a quadratic correction. A sympathetic reader would care because SG is a geophysical model derived from Euler, and these estimates make the SG-to-Euler limit quantitative in strong norms and on longer-than-hyperbolic time scales.

What carries the argument

The load-bearing object is the perturbative Monge-Ampère equation $\det(I+\varepsilon D^2 \psi) = 1 + \varepsilon \Delta \psi + \varepsilon^2 \det D^2 \psi$, which presents SG as a Poisson equation plus a quadratic determinant correction. The proof compares the SG and Euler Lagrangian flows, using three estimates: an endpoint logarithmic bound for the inverse Laplacian that controls $D^2 \psi$ in $L^\infty$ from the density in $C^\alpha$; a bilinear Jacobian estimate that places $\det D^2 \psi$ in $H^{-1}$ with norm controlled by $\| D^2 \psi \|_{L^2}^2$; and an $H^{-1}$ stability estimate bounding the gap of transported densities by the $L^2$ gap of their flows. These close a Grönwall inequality for the flow gap with an $O(\varepsilon)$ forcing term. The Wasserstein comparison then uses a deterministic flow repres

What would settle it

Look for a sequence of periodic solutions of $\det(I+\varepsilon D^2 \psi) = 1 + \varepsilon \rho$ on the torus, satisfying $\varepsilon \| \nabla \rho \|_{L^\infty} \le 1/4$ and $\| \rho \|_{C^\alpha}$ bounded, for which $\| D^2 \psi \|_{C^\alpha}$ is unbounded as $\varepsilon \to 0$; if such a sequence exists, the asserted Schauder estimate fails and the lifespan bootstrap collapses. Alternatively, numerically integrate $SG_\varepsilon$ and measure the first time the condition $\varepsilon \| \nabla \rho \|_{L^\infty} \le 1/4$ breaks, comparing that time with the predicted $\varepsilon^{-1} \log \log(1/\varepsilon)$ lower bound.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that the SG–Euler limit is not merely formal or weak: under the bootstrap condition $\varepsilon \| \nabla \rho^\varepsilon \|_{L^\infty} \le 1/4$, the semigeostrophic velocity differs from the Euler velocity by at most $C_t \varepsilon$ in $L^2$, and the physical densities differ by at most $C_t \varepsilon$ in the Wasserstein metric, while the bootstrap persists for slow time $\tau \sim \log \log(1/\varepsilon)$, i.e. physical time $\gtrsim \varepsilon^{-1} \log \log(1/\varepsilon)$. This is achieved by comparing Lagrangian flows rather than potentials, treating $\varepsilon \det D^2 \psi$ as an $H^{-1}$ perturbation, and using a logarithmic rather than quadratic growth law for the Hölder norm of the density.

Load-bearing premise

The load-bearing premise is the asserted Hölder ($C^\alpha$) Schauder estimate $\| D^2 \psi^\varepsilon \|_{C^\alpha} \le C_S \| \rho^\varepsilon \|_{C^\alpha}$ in Step 2 of Proposition 5.3, which is stated without proof; the text also passes from $\varepsilon \| \nabla \rho \|_{L^\infty} \le 1/4$ to a breakdown threshold $y \sim 1/\varepsilon$ with no connecting argument.

Editorial extensions

If this is right

  • If the central claim is right, semigeostrophic dynamics remain a strong quantitative stand-in for 2D Euler on the bootstrap window: the velocity error is O(ε) in L2, not merely in a weak topology.
  • The perturbative regime lasts at least ε^{-1} log log(1/ε) in physical time, strictly longer than the standard O(ε^{-1}) hyperbolic scale.
  • The physical densities m^ε and \bar m^ε are O(ε) close in the Wasserstein metric, so measure-level predictions of SG match Euler to first order.
  • Smooth initial data remain smooth on the bootstrap window, since the density's Hölder and Sobolev norms are controlled by the uniform Hessian bound.
  • A first-order corrector improves the velocity approximation to O(ε²), identifying the leading Euler correction explicitly and quantitatively.

Reading between the lines

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

  • The asserted but unproved Hölder (C^α) Schauder estimate in Step 2 of Proposition 5.3 is the natural point to attack: if it fails, both the log-log lifespan and the uniform Hessian control are jeopardized, so proving or disproving it would settle the paper's main theorems.
  • The Wasserstein comparison theorem is stated for SG–Euler but only uses one Lipschitz velocity, one rough velocity with an L^p moment bound, and the continuity-equation structure, so it likely transfers to any pair of transport equations satisfying those hypotheses.
  • Because log log(1/ε) grows extremely slowly, the numerical gain over 1/ε is modest for moderate ε; the structural significance is the replacement of quadratic growth by logarithmic growth in the bootstrap criterion, which could be tested by numerical simulation of the first breakdown time.
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.

Referee Report

4 major / 3 minor

Summary. The paper studies the two-dimensional semigeostrophic (SG) system on the torus in the small-amplitude scaling (2.2) and claims three outputs: a lifespan lower bound T*(ε) ≳ ε^{-1} log log(1/ε) on a bootstrap window, an O(ε) L^2 velocity comparison with 2D Euler on that window, and an O(ε) Wasserstein comparison of the physical densities, plus a second-order corrector result. The proof strategy is to treat ε det D^2ψ as a perturbation of the Poisson equation, combine endpoint Calderón–Zygmund, Wente, and Loeper-type H^{-1} stability estimates, and use an AGS superposition argument for the Wasserstein comparison. The central hinge is Proposition 5.3, which is supposed to give uniform C^α control of D^2ψ^ε in the bootstrap regime. I find that Proposition 5.3 is neither proved nor valid as stated, and that the lifespan inversion and the O(ε^2) theorem have additional structural gaps. The Wasserstein comparison framework in §7.2 is largely independent and appears plausible, but the advertised Theorems 4.1 and 4.2 are not established.

Significance. If the main results were correct, they would represent a meaningful advance: a log-log lifespan improvement over the standard ε^{-1} scale and a strong L^2 velocity convergence rate, improving Loeper's weak ε^{2/3} density convergence. The paper's use of AGS superposition for the Wasserstein comparison is a good idea and the comparison estimate (4.2) does not depend on the dubious elliptic regularity. The O(ε) consequence, however, depends on Theorem 4.2, which in turn depends on Proposition 5.3. No fitting parameters are introduced and the estimates are derived from external tools, but the central elliptic regularity step is asserted rather than proved, and part of the supporting algebra is false. The manuscript is therefore not acceptable in its present form.

major comments (4)
  1. [§5, Proposition 5.3, Step 2] The asserted bound ||D^2ψ^ε||_{C^α} ≤ C_S Z is the hinge of both Theorem 4.1 and Theorem 4.2, but it is not proved and does not follow from the displayed equation. Linear Schauder applied to Δψ^ε = ρ^ε - ε det D^2ψ^ε gives ||D^2ψ^ε||_{C^α} ≤ C(||ρ^ε||_{C^α} + ε||det D^2ψ^ε||_{C^α}); the second term needs ||D^2ψ^ε||_{C^α} itself. No nonlinear Monge–Ampère C^{2,α} estimate with constants depending only on α, T^2, M_0 and the bootstrap bound is stated or cited. The sentence 'Standard Schauder estimates ... imply' is circular: the bootstrap condition (5.4) controls ∇ρ^ε, not D^2ψ^ε.
  2. [§5, Proposition 5.3, Step 1] The chain ε||D^2ψ^ε||_{L∞}^2 ≲ ε||ρ^ε||_{L∞}^{2(1-α)}||∇ρ^ε||_{L∞}^{2α} ≲ M_0 is invalid under (5.4). The bootstrap assumption gives ε||∇ρ^ε||_{L∞} ≤ 1/4, so the right-hand side is at most ε^{1-2α}(1/4)^{2α}||ρ^ε||_{L∞}^{2(1-α)}, which diverges as ε→0 unless α=1/2. In addition, the text uses 'the bootstrap condition εY≲1/4' with Y=||D^2ψ^ε||_{L∞}, which is exactly the quantity to be proved bounded. Consequently the estimate ||F||_{L∞} ≲ M_0 in (5.7) has no valid derivation.
  3. [§6, proof of Theorem 4.1] The lifespan inversion has an unjustified threshold. The bootstrap regime is defined by ε||∇ρ^ε||_{L∞} ≤ 1/4 in (5.4), but the proof says the regime breaks down when y(τ*) = ||ρ^ε||_{C^α} is of order 1/ε. No estimate connects the C^α norm to the L∞ norm of the gradient, and the displayed 'Specifically, y(τ*)≈ c/ε' is an assumption, not a consequence of (4.1). Thus even if Proposition 5.3 were valid, the derivation T*(ε)≥(ε C* M_0)^{-1} log log(1/ε) does not follow as written.
  4. [§8.1, Theorem 8.4] The initial data are incompatible. The theorem assumes ρ^ε(0)=barρ(0)=ρ_0 and ψ^ε(0)=barφ(0). The SG elliptic equation (2.2) at t=0 then requires Δψ^ε(0)=ρ^ε(0)-ε detD^2ψ^ε(0), i.e. Δbarφ(0)=ρ_0-ε detD^2barφ(0). Since Δbarφ(0)=ρ_0 by the Euler equation, this forces detD^2barφ(0)=0, which is not true for generic smooth ρ_0. Hence the O(ε^2) corrector theorem cannot hold for the claimed class of data.
minor comments (3)
  1. [Abstract] The abstract and the full text contradict each other on the Wasserstein rate: the abstract advertises an O(ε^2) Wasserstein comparison, while Corollary 4.4 proves only W_2 ≤ C_t ε. The abstract also advertises a Lie–Poisson Hamiltonian expansion and cubic Monge–Ampère functional that do not appear anywhere in the body.
  2. [§2 and Theorem 4.2] Theorem 4.2 says both equations start 'with the same initial datum ρ_0', but the SG system (2.2) requires a compatible initial stream function ψ^ε(0) solving Δψ^ε(0)=ρ_0-ε detD^2ψ^ε(0). The statement should specify how ψ^ε(0) is chosen; otherwise the comparison of flows and velocities is not fully defined.
  3. [§7.2, Remark 7.5] The bound |∇P^ε - x| ≤ diam(T^2) is asserted without derivation. It should be justified from the optimal-transport structure or replaced by a precise reference, since it is used to verify the integrability hypothesis of the superposition theorem.

Circularity Check

0 steps flagged · score 1.0 of 10

No constructional circularity; the main theorems rest on an unproved C^alpha Schauder assertion and a threshold mismatch, which are rigor gaps rather than circular reductions.

full rationale

No fitted parameter is renamed as a prediction, and no load-bearing self-citation appears (the author cites no prior work of their own). The external ingredients — Loeper's flow-to-field H^{-1} stability, Wente-type Jacobian estimates, the endpoint Calderón-Zygmund bound, and the AGS superposition theorems — are genuinely independent and are not derived from the conclusions. The most delicate step is Proposition 5.3, Step 2: 'Standard Schauder estimates for the (perturbative) Monge-Ampère equation in the bootstrap regime imply ||D^2ψ^ε||_{C^α}≤C_S Z', where Z=||ρ^ε||_{C^α}. This estimate is asserted without proof or citation and is used to bound ||F||_{C^α}, so it is load-bearing for both Theorem 4.1 and Theorem 4.2. If the C^{2,α} regularity is not available with ε-independent constants, the endpoint CZ closure fails. That is an omitted proof, not a reduction of the conclusion to its own input. In addition, Section 6 states that the bootstrap 'breaks down when y(τ) reaches a threshold of order O(1/ε)' even though the actual bootstrap condition (4.1) is ε||∇ρ^ε||_{L∞}≤1/4; no argument links y=||ρ||_{C^α} to the gradient threshold. These are correctness/rigor gaps, not constructional circularity, so the circularity score is 1 rather than 0.

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

No fitted constants or new entities are introduced; the paper uses external PDE/OT tools. The main 'free choices' are universal bootstrap constants and the Hölder exponent α, neither fitted to data. The load-bearing burden is in unproved regularity assumptions, not in fitted parameters.

assumptions (6)
  • standard math Brenier polar factorization and periodic Monge–Ampère representation
    Section 3.1 assumes the dual SG formulation with det(I+D^2ψ)=ρ and convex potential; standard in [Bre91, Loe06].
  • standard math Loeper's H^{-1} flow stability estimate (Prop 3.1 / (5.3))
    Used to convert flow gaps to density/velocity gaps; proof in Prop 3.1 is a variant of Loeper Prop 4.2, but its own proof is abbreviated.
  • standard math Endpoint Calderón–Zygmund bound (5.1)
    Cited to [MBO02]; gives ||D^2(-∆)^{-1}f||∞ ≤ C_α||f||∞(1+log^+(||f||_{C^α}/||f||∞)).
  • domain assumption Wente-type bound for Hessian determinant
    Lemma A.1 is stated for R^2 ([DP17]); applied on T^2 in (5.2) without a periodization argument.
  • ad hoc to paper C^α Schauder estimate for the perturbed Monge–Ampère/Poisson equation
    Prop 5.3 Step 2 asserts ||D^2ψ^ε||_{C^α}≤C_S∥ρ∥_{C^α} without proof or citation; load-bearing for the bootstrap closure.
  • ad hoc to paper Initial compatibility ψ^ε(0)=\barϕ(0) in Theorem 8.4
    Incompatible with ∆ψ^ε(0)=ρ0−εdetD^2ψ^ε(0) when ρ^ε(0)=ρ0; introduces an O(ε) initial velocity error that the O(ε^2) claim ignores.

how reviews work

0 comments
Cite this review

Pith. "Pith review of The Semigeostrophic--Euler Limit via Perturbative Monge--Amp\`ere Estimates." pith.science (2026). https://pith.science/paper/OTVKUGBP

@misc{pith2026260104797,
  author       = {Pith},
  title        = {Pith review of: The Semigeostrophic--Euler Limit via Perturbative Monge--Amp\`ere Estimates},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OTVKUGBP}},
  note         = {Machine review of arXiv:2601.04797}
}
abstract

We study the two-dimensional semigeostrophic system on the flat torus in the small-amplitude regime. We formulate the rescaled dynamics as the Lie--Poisson flow of a renormalized optimal-transport energy and expand this Hamiltonian in \(C^1\). The leading term is the Euler Hamiltonian, while the first correction is an explicit cubic Monge--Amp\`ere functional. We then derive quantitative consequences for the semigeostrophic--Euler limit: a perturbative scale-uniform endpoint Monge--Amp\`ere estimate under Hessian pinching, an explicit logarithmic perturbative lifespan for the strong branch, fixed-slow-time \(O(\eps)\) velocity convergence for canonically prepared strong branches, a conditional weak--strong rate-transfer corollary, and an \(O(\eps^2)\) Wasserstein comparison for the physical densities.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

3 extracted references · 1 linked inside Pith

  1. [1]

    Gradient flows with metric and differentiable structures, and applications to the Wasserstein space

    [AGS04] Luigi Ambrosio, Nicola Gigli, and Giuseppe Savaré. “Gradient flows with metric and differentiable structures, and applications to the Wasserstein space”. In:Atti della Accademia Nazionale dei Lincei. Classe di Scienze Fisiche, Matematiche e Naturali. Rendiconti Lincei. Matematica e Applicazioni15.3-4 (2004), pp. 327–

    2004

  2. [343]

    A global existence result for the semigeostrophic equations in three dimensional convex domains

    [Amb+12] Luigi Ambrosio et al. “A global existence result for the semigeostrophic equations in three dimensional convex domains”. In:arXiv preprint arXiv:1205.5435(2012). [BC09] Yann Brenier and Mike Cullen. “Rigorous derivation of the XZ semigeostrophic equations”. In: (2009). [Bre91] Yann Brenier. “Polar factorization and monotone rearrangement of vecto...

    arXiv 2012

  3. [2014]

    A fully nonlinear version of the incompressible Euler equations: the semigeostrophic system

    Springer, 2018, pp. 1–42. [Loe06] Grégoire Loeper. “A fully nonlinear version of the incompressible Euler equations: the semigeostrophic system”. In:SIAM journal on mathematical analysis38.3 (2006), pp. 795–823. [MBO02] Andrew J Majda, Andrea L Bertozzi, and A Ogawa. “Vorticity and incompressible flow. Cambridge texts in applied mathematics”. In:Appl. Mec...

    2018

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.