Inviscid limit to the shock waves for the fractal Burgers equation

Alexis Vasseur; Moon-Jin Kang; Sona Akopian

REVIEW 1 major objections 4 minor 27 references

Inviscid limit to the shock waves for the fractal Burgers equation

T0 review · 1 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 Entropy shocks of the fractal Burgers equation are stable in the vanishing-viscosity limit with a quantified rate, even for large perturbations.

desk verdict Genuine extension of the relative entropy method to the fractional Laplacian, but a real inequality slip in the final step undermines the stated rate. read the letter →

arxiv 1908.01632 v2 pith:NGJWWRR5 submitted 2019-08-05 math.AP

Sona Akopian , Moon-Jin Kang , Alexis Vasseur This is my paper · ORCID

classification math.AP MSC 35L6535L6735B3535B40

keywords fractalBurgersequationfractionalLaplacianinviscidlimitentropyshockrelativemethodlayerlargeperturbationvanishingviscosity

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 proves the first quantitative vanishing-viscosity limit for entropy shocks of the fractal Burgers equation $u_t + (u^2/2)_x = \varepsilon \Delta^{\alpha/2}u$ in the range $1<\alpha<2$. It shows that for any initial datum that is an $L^2$ perturbation of an entropy shock and whose derivative has a positive part in $L^2$, the viscous solution, after shifting the spatial coordinate by a Lipschitz function $X_\varepsilon(t)$, stays within distance $\|u_0-S_0\|_{L^2} + C(T)\psi(\varepsilon)$ of the inviscid shock on any bounded time interval, with $\psi(\varepsilon)\to0$ as $\varepsilon\to0$. The significance is that the perturbation may be large, so the shock layer is destroyed in the limit, yet the entropy shock is still recovered. The rate $\psi(\varepsilon)$ is explicit in terms of the dissipation parameters and the tail of the viscous shock layer, though the layer tail's own decay rate is left open.

What carries the argument

The mechanism is the relative entropy method with a localized weight. The paper tracks $H(t)=\int_{\mathbb{R}}\phi_\delta^2(|x|/\varepsilon^\beta)\frac{|u_\varepsilon(x+X(t),t)-S_\varepsilon(x)|^2}{2}\,dx$, where $S_\varepsilon(x)=S_1(x/\varepsilon^\beta)$ is the monotone shock layer of the scaled equation, $\phi_\delta$ is a smooth cutoff that localizes to a window of width $\delta\varepsilon^\beta$, and the shift $X(t)$ is chosen to solve $\dot X=f(u_\varepsilon(X(t),t),S_\varepsilon(0))$, with $f$ the normalized relative entropy flux from [20]. The choice of shift cancels the main hyperbolic term; the remaining three contributions are bounded by $C\sqrt{\delta\varepsilon^\beta}$ (hyperbolic term from the cutoff), by $C((1/\delta)^{3/2}+(S_1(\sqrt\delta)-u_+)+(u_- - S_1(-\sqrt\delta)))$ (a second hyperbolic term controlled by the monotonicity of the layer), and by $C(1/\delta)^{\alpha-1}$ (the fractional dissipation term). Minimizing over $\delta\ge4$ yields $\psi(\varepsilon)$, and the layer's monotonicity, proved from the strict convexity of the flux and the structure of the fractional Laplacian, is what makes the middle estimate possible.

What would settle it

Take $\alpha=3/2$, $u_-=1$, $u_+=-1$, and initial data $u_0=S_0+\chi_{[-1,1]}$ after rescaling so that $\|u_0-S_0\|_{L^2}=1$; compute numerically, at $t=1$ and $\varepsilon=10^{-2},10^{-3},\ldots$, the $L^2$ error between $u_\varepsilon(\cdot+X(t),t)$ and $S_0$ using the shift defined by the paper's ODE. If the quantity $\|u_\varepsilon(\cdot+X(t),1)-S_0\|_{L^2}-\|u_0-S_0\|_{L^2}$ does not tend to zero as $\varepsilon\to0$, Theorem 1.1 is false.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central result is Theorem 1.1: for $1<\alpha<2$, any end states $u_->u_+$, and initial data $u_0$ with $u_0-S_0\in L^2(\mathbb{R})$ and $(\partial_x u_0)_+\in L^2(\mathbb{R})$, for every $T>0$ there exists a Lipschitz shift $X_\varepsilon$ with $X_\varepsilon(0)=0$ such that for all $t\le T$, \[\|u_\varepsilon(\cdot+X_\varepsilon(t),t)-S_0(\cdot-\$\sigma$ t)\|_{$L^{2}$(\mathbb{R})}\le \|u_0-S_0\|_{$L^{2}$(\mathbb{R})}+C(T)\psi(\varepsilon),\] where $\psi(\varepsilon)\to0$. This is the first result of this kind for the fractal Burgers equation: previous convergence statements either used no initial perturbation or restricted to small perturbations, whereas here the perturbation can be arbitrarily large in $L^2$. A shift is necessary, since $L^2$ contraction to the shock without a shift is already known to fail. The proof establishes the same estimate for convergence to the smooth monotone shock layer $S_\varepsilon(x)=S_1(x/\varepsilon^\beta)$ of width $\varepsilon^\beta$, and then uses the $L^2$ convergence of the layer to the discontinuous shock.

Load-bearing premise

The proof needs a smooth monotone shock layer $S_1$ for the stationary fractional Burgers equation, and the claimed rate inherits the unknown rate at which that layer approaches the end states $u_\pm$; if the layer tail decays very slowly or the layer is not monotone, the estimate loses its rate or its control of the hyperbolic term.

Editorial extensions

If this is right

For every $1<\alpha<2$ and every bounded time interval, the entropy shock is stable in $L^2$ under the vanishing-viscosity evolution for large perturbations, with the spatial shift absorbing the phase error that would otherwise make contraction fail.
When the shock layer decays exponentially, the excess error is at most $C\varepsilon^{1/(2(2\alpha-1))}$, a rate only slightly worse than the optimal layer rate $\varepsilon^{1/(2(\alpha-1))}$.
The argument is written for a general strictly convex flux $A$, so any scalar conservation law with a convex nonlinearity and a monotone fractional shock layer inherits the same quantitative stability statement.
If the initial datum is exactly the scaled layer, the rate matches the optimal layer rate, showing the theorem is sharp up to the unresolved layer-tail contribution.

Reading between the lines

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

The shift $X_\varepsilon$ is not a nuisance: it carries the phase drift caused by the excess entropy of a large perturbation, and the same shift-based cancellation may be the right framework for other nonlocal conservation laws with monotone traveling waves.
A sharper closed-form estimate for the tail of $S_1$ at specific $\alpha$ would immediately upgrade $\psi(\varepsilon)$ to an explicit power of $\varepsilon$; this is the natural next computation.
The hypothesis $(\partial_x u_0)_+\in L^2$ is a one-sided regularity condition that could perhaps be relaxed, but testing large oscillatory data with unbounded positive slope would show whether the method's reliance on this bound is essential.
As $\alpha\to2^-$, $\beta=1/(\alpha-1)\to\infty$, so the layer width is extremely small; comparing the rate with the local Laplacian case suggests the large-perturbation statement is most delicate near $\alpha=2$.

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

Good to see this paper out. You should know two things. The main advance is real: it genuinely extends Choi–Vasseur to the nonlocal fractional Laplacian for large L2 perturbations, and the nonlocal parabolic estimate (Prop. 2.3) is solid new work. But the last estimate in Section 2.6 has a genuine bug: they move from sqrt(δεβ + E) to sqrt(δεβ) + E, which fails when E is small and positive. Since E is exactly the small parameter that tends to zero, the stated ψ(ε) is not actually derived. That doesn't kill the argument: the valid inequality sqrt(a+E) ≤ sqrt(a)+sqrt(E) gives a different still-vanishing rate with square roots of the layer-tail terms. So it's a patch, not a hole in the strategy.

The rest of the proof holds up. Lemma 2.1 (monotonicity of the layer) and Lemma 2.2 (one-sided derivative bound) are clean. The cutoff argument with φδ is careful. The authors are honest in Remark 1.1 that the layer decay rate is unknown, which means the 'quantified' rate is not a clean power law; the abstract oversells it a bit. Self-citations to Choi–Vasseur and Kang–Vasseur are legitimate; they are prior results, not a fit.

My take: this deserves a serious referee and likely publication after a straightforward correction. The flawed step is easy to fix, and the main theorem still holds in a slightly modified form. I'd send it to review rather than desk reject, with the instruction to fix the sqrt inequality and make the rate claim match what is actually proven.

For your reading group, it's worth a session: the nonlocal estimates are instructive, and the final slip is a nice lesson in why inequality steps need checking.

Referee Report

1 major / 4 minor

Summary. The paper studies the vanishing fractal-viscosity limit for the one-dimensional fractal Burgers equation ∂_t u + ∂_x(u^2/2) = ε Δ^{α/2}u with 1 < α < 2. The main result, Theorem 1.1, asserts an L^2 convergence estimate, up to a Lipschitz shift X_ε(t), from solutions of the scaled equation to the inviscid entropy shock S_0, for initial data that are L^∞ with u_0 - S_0 ∈ L^2 and (u_0')_+ ∈ L^2. The rate is encoded in a function ψ(ε) built from the shock layer S_1 of the stationary profile. The proof combines the relative entropy method with a shift ODE taken from Leger's normalized relative flux, a monotonicity lemma for the layer, a one-sided derivative bound, and a nonlocal parabolic estimate.

Significance. If the stated result holds, it would be a valuable extension of the relative-entropy inviscid-limit theory to nonlocal viscosity for large perturbations, and several components of the proof are correct and useful: the monotonicity of the shock layer in Lemma 2.1, the positive-part derivative bound in Lemma 2.2, the shift construction, and the nonlocal parabolic estimate in Proposition 2.3. However, the final rate as stated is not established because of an invalid inequality in the concluding step, so the paper requires a nontrivial correction before the central quantitative claim can be accepted.

major comments (1)

[§2.6, final display; Theorem 1.1] The proof obtains the valid bound ‖u_ε(·+X(t),t)-S_ε‖_{L^2} ≤ ‖u_0-S_ε‖_{L^2} + C(T)√(δε^β+E(ε,δ)), where E(ε,δ)=√(δε^β)+(S_1(√δ)-u_+)+(u_- - S_1(-√δ))+(1/δ)^{α-1}. The next displayed step replaces C(T)√(δε^β+E) by C(T)(√(δε^β)+E). This uses the inequality √(a+b) ≤ √a + b, which is false whenever 0 < b < 1. Since Remark 1.2 chooses δ=ε^{-β/2} precisely to make E(ε,δ)→0, the inequality fails in exactly the regime needed for convergence. The formula for ψ(ε) displayed in Theorem 1.1, with two linear √(δε^β) terms and E appearing linearly, is precisely the outcome of this invalid step, so the stated rate is not derived. The error is repairable: using the valid bound √(δε^β+E) ≤ √(δε^β)+√E would give a still-vanishing rate, with square roots of the layer-tail terms, and would preserve the qualitative conclusion. But as written, the central quantitative claim of the paper is not proved.

minor comments (4)

[Theorem 1.1] The definition of ψ(ε) refers to E(ε,δ) even though E is not defined until §2.6, and the displayed expression with two √(δε^β) terms is inconsistent with the proof's final formula. The theorem statement should define all quantities explicitly and match the bound actually proved.
[Remark 1.1; Theorem 1.1] The word 'quantified' in the abstract and Theorem 1.1 is stronger than what is delivered, since ψ(ε) depends on the layer-tail values S_1(√δ)-u_± whose decay rate is unknown and is explicitly said to be unknown in Remark 1.1. The authors should rephrase the claim, for example as 'a rate expressible in terms of the layer tail', unless an independent estimate for that tail is supplied.
[§2.2 and §2.4] The notation for the localization weight is inconsistent across sections: Proposition 2.1 writes φ_δ(|x|/ε^β), while after the change of variables in Proposition 2.2 the weight is written φ_δ(|x|). Defining the rescaled weight once and then referring to it would improve readability.
[§2.3, proof of Proposition 2.1] In the estimate of h_1, the notation v_ε(0,t) = u_ε(X(t),t) is used without restating the definition from (2.13); a short reminder would help the reader follow the shift-dependent argument.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the central estimates are proved in-paper, and the self-citations are methodological rather than load-bearing.

full rationale

The paper's main theorem is not obtained by fitting a parameter to the target result or by defining the conclusion into the hypotheses. The inviscid-limit bound (1.7) is derived through the relative entropy method: Lemma 2.3 computes H'(t), Propositions 2.1-2.3 bound the hyperbolic and nonlocal terms, and Section 2.6 assembles the estimate. The shift X is constructed as the solution to the ODE (2.17), whose existence is proved via Cauchy-Lipschitz and Lemma 2.4; it is not a free parameter fitted to force the inequality. The principal external inputs are Chmaj [5] for the existence of the shock layer S1 and Droniou-Gallouet-Vovelle [10] for global well-posedness and L-infinity stability; neither is by the present authors, and neither contains the theorem being proved. The self-citations, e.g., [6] and [16], supply the relative-entropy technique and algebraic identities, but the paper reproduces the needed estimates (Lemmas 2.1 and 2.2) and provides the new nonlocal estimates (Propositions 2.2 and 2.3) in full; the load-bearing arguments do not reduce to an unverified self-citation chain. The admitted limitation that 'the rate of convergence of the shock layer to the two end states u± is not known' means that the rate psi inherits layer-tail data; this is a caveat on explicitness, not a circular reduction. Likewise, the skeptical point about a possible invalid inequality involving square roots in Section 2.6 is a correctness concern about the final estimate, not a circularity: it does not make the theorem's statement equivalent to its inputs by construction. Overall, there is no significant circularity; the small score reflects the presence of several self-citations in the method, but they are not load-bearing.

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

No free parameters are fitted; the proof introduces no new physical entities. The load-bearing inputs are the cited well-posedness theory for the fractional Burgers equation and the existence of the shock layer. The cutoff function and the shift X_epsilon are mathematical constructions, not invented entities.

assumptions (4)

domain assumption Global existence, uniqueness and L-infinity maximum principle for the scaled fractal Burgers equation (2.1) for L-infinity initial data, 1 < alpha < 2.
Invoked in Section 2, after equation (2.1), citing Droniou-Gallouet-Vovelle [10]. The proof uses the bound ||u_epsilon||_{L-infinity} <= ||u0||_{L-infinity} to control the relative entropy terms.
domain assumption Existence of a smooth, monotone shock layer S1 solving the stationary traveling wave equation (2.2) with limits u± as x goes to ±infinity.
Invoked in Theorem 1.1 and Section 2.1, citing Chmaj [5]. The monotonicity is proved in Lemma 2.1 but the existence and limits are used as prior results.
standard math The normalized relative entropy flux f(u,v) satisfies the bounds in Lemma 2.4, taken from Leger [20].
Used in the proof of Proposition 2.1 and the definition of the shift ODE (2.17). This is a cited standard lemma for convex scalar conservation laws.
standard math The cutoff function phi in (2.6) and (2.19) is smooth, nondecreasing, and satisfies the stated endpoint values.
The specific form (2.19) is used in Proposition 2.2 to get the (1/delta)^{3/2} bound. This is a construction, not a physical assumption.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Inviscid limit to the shock waves for the fractal Burgers equation." pith.science (2026). https://pith.science/paper/NGJWWRR5

@misc{pith2026190801632,
  author       = {Pith},
  title        = {Pith review of: Inviscid limit to the shock waves for the fractal Burgers equation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NGJWWRR5}},
  note         = {Machine review of arXiv:1908.01632}
}

abstract

We show the vanishing viscosity limit to entropy shocks for the fractal Burgers equation in one space dimension. More precisely, we quantify the rate of convergence of the inviscid limit in $L^2$ for large initial perturbations around the entropy shock on any bounded time interval. This is the first result on the inviscid limit to entropy shock for the fractal Burgers equation with the quantified convergence, for large initial perturbations.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 27 canonical work pages

[1]

N. Alibaud. Entropy formulation for fractal conservati on laws. J. Evol. Equ. , 7:145–175, 2007

work page 2007
[2]

Alibaud, C

N. Alibaud, C. Imbert, and G. Karch. Asymptotic properti es of entropy solutions to fractal Burgers equation,. SIAM J. Math. Anal. , 42:354–376, 2010

work page 2010
[3]

Biler, T

P. Biler, T. Funaki, and W. Woyczynski. Fractal Burgers e quations. J. Diﬀerential Equations , 148:9–46, 1998

work page 1998
[4]

C. H. Chan and M. Czubak. Regularity of solutions for the c ritical N-dimensional Burgers equation,. Ann. Inst. H. Poincar´ e Anal. Non Lin´ eaire, 27:471–501, 2010

work page 2010
[5]

A. Chmaj. Existence of travelling waves in the fractiona l Burgers equation. Bull. Aus. Math. Soc. , 97:102–109, 2018

work page 2018
[6]

Choi and A

K. Choi and A. Vasseur. Short-time stability of scalar vi scous shocks in the inviscid limit by the relative entropy method. SIAM J. Math. Anal. , 47:1405–1418, 2015

work page 2015
[7]

C. M. Dafermos. The second law of thermodynamics and stab ility. Arch. Rational Mech. Anal. , 70(2):167–179, 1979. 15

work page 1979
[8]

R. J. DiPerna. Uniqueness of solutions to hyperbolic con servation laws. Indiana Univ. Math. J. , 28(1):137–188, 1979

work page 1979

Show all 27 references

[9]

J. Droniou. Vanishing non-local regularization of a sca lar conservation law. Electron. J. Diﬀerential Equations, 117:1–20, 2003

2003
[10]

Droniou, T

J. Droniou, T. Gallouet, and J. Vovelle. Global solutio n and smoothing eﬀect for a non-local regular- ization of a hyperbolic equation. J. Evol. Equ. , 3:499–521, 2002

2002
[11]

M.-J. Kang. L2-type contraction for shocks of scalar viscous conservatio n laws with strictly convex ﬂux. https://arxiv.org/pdf/1901.02969.pdf

1901 arXiv
[12]

M.-J. Kang. Non-contraction of intermediate admissib le discontinuities for 3-D planar isentropic mag- netohydrodynamics. Kinet. Relat. Models , 11(1):107–118, 2018

2018
[13]

Kang and A

M.-J. Kang and A. Vasseur. Contraction property for lar ge perturbations of shocks of the barotropic Navier-Stokes system. J. Eur. Math. Soc. (JEMS), To appear. https://arxiv.org/pd f/1712.07348.pdf

arXiv
[14]

Kang and A

M.-J. Kang and A. Vasseur. Uniqueness and stability of e ntropy shocks to the isentropic Euler system in a class of inviscid limits from a large family of Navier-Stokes systems. https://arxiv.org/pdf/1902.01792.pdf

1902 arXiv
[15]

Kang and A

M.-J. Kang and A. Vasseur. Criteria on contractions for entropic discontinuities of systems of conser- vation laws. Arch. Ration. Mech. Anal. , 222(1):343–391, 2016

2016
[16]

Kang and A

M.-J. Kang and A. Vasseur. L2-contraction for shock waves of scalar viscous conservatio n laws. Annales de l’Institut Henri Poincar´ e (C) : Analyse non lin´ eaire, 34(1):139156, 2017

2017
[17]

M.-J. Kang, A. Vasseur, and Y. Wang. L2-contraction for planar shock waves of multi-dimensional scalar viscous conservation laws. J. Diﬀerential Equations , 267:2737–2791, 2019

2019
[18]

Karch, C

G. Karch, C. Miao, and X. Xu. On convergence of solutions of fractal Burgers equation toward rarefac- tion waves,. SIAM J. Math. Anal. , 39:1536–1549, 2008

2008
[19]

Kiselev, F

A. Kiselev, F. Nazarov, and R. Shterenberg. Blow up and r egularity for fractal Burgers equation. Dyn. Partial Diﬀer. Equ. , 5:211–240, 2008

2008
[20]

N. Leger. L2 stability estimates for shock solutions of scalar conserva tion laws using the relative entropy method. Arch. Ration. Mech. Anal. , 199(3):761–778, 2011

2011
[21]

D. Serre. Systems of conservation laws I, II . Cambridge University Press, Cambridge, 1999

1999
[22]

Serre and A

D. Serre and A. Vasseur. L2-type contraction for systems of conservation laws. J. ´Ec. polytech. Math. , 1:1–28, 2014

2014
[23]

Serre and A

D. Serre and A. Vasseur. About the relative entropy meth od for hyperbolic systems of conservation laws. Contemp. Math. AMS , 658:237–248, 2016

2016
[24]

Serre and A

D. Serre and A. Vasseur. The relative entropy method for the stability of intermediate shock waves; the rich case. Discrete Contin. Dyn. Syst. , 36(8):4569–4577, 2016

2016
[25]

A. Vasseur. Recent results on hydrodynamic limits. In Handbook of diﬀerential equations: evolutionary equations. Vol. IV , Handb. Diﬀer. Equ., pages 323–376. Elsevier/North-Holla nd, Amsterdam, 2008

2008
[26]

A. Vasseur. Relative entropy and contraction for extre mal shocks of conservation laws up to a shift. In Recent advances in partial diﬀerential equations and applic ations, volume 666 of Contemp. Math. , pages 385–404. Amer. Math. Soc., Providence, RI, 2016

2016
[27]

Vasseur and Y

A. Vasseur and Y. Wang. The inviscid limit to a contact di scontinuity for the compressible navier- stokes-fourier system using the relative entropy method. SIAM J. Math. Anal. , 47(6):4350–4359, 2015. (Sona Akopian) Division of Applied Mathematics, Brown University E-mail add...

2015

Pith tools

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

[1] [1]

N. Alibaud. Entropy formulation for fractal conservati on laws. J. Evol. Equ. , 7:145–175, 2007

work page 2007

[2] [2]

Alibaud, C

N. Alibaud, C. Imbert, and G. Karch. Asymptotic properti es of entropy solutions to fractal Burgers equation,. SIAM J. Math. Anal. , 42:354–376, 2010

work page 2010

[3] [3]

Biler, T

P. Biler, T. Funaki, and W. Woyczynski. Fractal Burgers e quations. J. Diﬀerential Equations , 148:9–46, 1998

work page 1998

[4] [4]

C. H. Chan and M. Czubak. Regularity of solutions for the c ritical N-dimensional Burgers equation,. Ann. Inst. H. Poincar´ e Anal. Non Lin´ eaire, 27:471–501, 2010

work page 2010

[5] [5]

A. Chmaj. Existence of travelling waves in the fractiona l Burgers equation. Bull. Aus. Math. Soc. , 97:102–109, 2018

work page 2018

[6] [6]

Choi and A

K. Choi and A. Vasseur. Short-time stability of scalar vi scous shocks in the inviscid limit by the relative entropy method. SIAM J. Math. Anal. , 47:1405–1418, 2015

work page 2015

[7] [7]

C. M. Dafermos. The second law of thermodynamics and stab ility. Arch. Rational Mech. Anal. , 70(2):167–179, 1979. 15

work page 1979

[8] [8]

R. J. DiPerna. Uniqueness of solutions to hyperbolic con servation laws. Indiana Univ. Math. J. , 28(1):137–188, 1979

work page 1979