Classification of the blow-up behavior for a semilinear wave equation with nonconstant degenerate coefficients

Asma Azaiez; Hatem Zaag

REVIEW 2 major objections 4 minor 16 references

Classification of the blow-up behavior for a semilinear wave equation with nonconstant degenerate coefficients

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 Blow-up profiles persist for degenerate wave equations

desk verdict The variable-coefficient Lyapunov functional is a genuine addition, but the stated blow-up profiles drop the b(x0)^{-1/(p-1)} factor and the main theorems are false as written. read the letter →

arxiv 1908.02081 v2 pith:F6Q4CRZF submitted 2019-08-06 math.AP

Asma Azaiez , Hatem Zaag This is my paper · ORCID

classification math.AP MSC 35L0535L7135B4435B40

keywords semilinearwaveequationnonconstantcoefficientsdegenerateblow-upprofilecharacteristicpointssimilarityvariablessolitoncurveregularity

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 aims to establish that the blow-up classification known for constant-coefficient semilinear wave equations survives when the wave speed is nonconstant and can degenerate at the origin. A change of variables $X=\varphi(x)=\int_0^x \frac{dy}{\sqrt{a(y)}}$ reduces the radial equation to a constant-coefficient wave equation perturbed by a nonconstant reaction coefficient $\beta(X)$, and the paper shows that the standard blow-up classification extends to this setting. Outside the origin, non-characteristic blow-up points have a universal single-soliton profile with exponential convergence, while characteristic points relax to an alternating sum of solitons and are isolated. At the origin, where the degeneracy acts, the paper proves only boundedness in similarity variables and leaves the full profile classification open.

What carries the argument

The load-bearing object is the change of variables $X=\varphi(x)=\int_0^x \frac{dy}{\sqrt{a(y)}}$, which converts the variable-speed equation into the constant-coefficient form $\partial_t^2 U = \partial_X^2 U + \frac{d-1}{X}\partial_X U + \beta(X)|U|^{p-1}U + f(U)+G$, with only the reaction coefficient $\beta$ genuinely nonconstant. On this reduced equation one uses the self-similar variables $w_{x_0}(y,s)=(T(x_0)-t)^{2/(p-1)}u(x,t)$, $y=(\varphi(x)-\varphi(x_0))/(T(x_0)-t)$, $s=-\log(T(x_0)-t)$, together with the explicit soliton family $\kappa(\hat d,y)=\kappa_0(1-\hat d^2)^{1/(p-1)}(1+\hat d y)^{-2/(p-1)}$. The asymptotic conclusions are carried by a Lyapunov functional $E=E_0+I+J+K$ built in Section 2.1 for the perturbed equation; its monotonicity, combined with the covering argument for lower bounds and the spectral analysis imported from the constant-coefficient theory, drives convergence to the soliton or multisoliton profile.

What would settle it

Take equation (1.12) with a strongly nonconstant $\beta$, for instance $\beta(X)=1+\varepsilon\sin X$, and compute the similarity-variable solution at a non-characteristic blow-up point; if the distance to $\theta\kappa(T'\sqrt a,\cdot)$ in $H^1_\rho\times L^2_\rho$ fails to decay exponentially, or if the rate constant depends on $\varepsilon$, the claimed transfer from the constant-coefficient case is false. On the characteristic side, verify numerically or asymptotically whether the center formula (1.26) and the logarithmic asymptotics (1.27)-(1.28) hold with constants $\nu(p)$, $c_1(p)$ independent of $\beta$.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that after the change of variables $X=\varphi(x)=\int_0^x \frac{dy}{\sqrt{a(y)}}$ and passage to similarity variables, every blow-up point $x_0\neq 0$ obeys the same dichotomy as in the constant-coefficient case. For a non-characteristic point, $(w,\partial_s w)$ converges in $H^1_\rho\times L^2_\rho$ to $\theta(x_0)\kappa(T'(x_0)\sqrt{a(x_0)},\cdot)$ with exponential rate $e^{-\mu_0(s-s^*)}$, and in the original variables $u(x,t)$ is asymptotic to $\theta(x_0)\kappa_0(1-a(x_0)|T'(x_0)|^2)^{1/(p-1)}(T(x_0)-t+T'(x_0)\sqrt{a(x_0)}(\varphi(x)-\varphi(x_0)))^{-2/(p-1)}$. For a characteristic point, the limit is a sum of $k(x_0)\ge 2$ alternating solitons with explicitly spaced centers, and such points are isolated, with sharp logarithmic asymptotics for $T'$ and $T$ nearby. At the origin, the paper obtains a uniform bound in the weighted spaces $H^1_{\rho_0}\times L^2_{\rho_0}$ for non-characteristic data, but does not claim the full profile classification there.

Load-bearing premise

The paper's main theorems are not proved in detail; they are asserted to follow from prior works by 'very minor adaptations', so the load-bearing premise is that the complete blow-up classification machinery—the Lyapunov monotonicity, the covering argument, the spectral gap, and the isolatedness proof for characteristic points—transfers without essential change to the nonconstant-coefficient equation $\beta(X)\not\equiv 1$, including at the degenerate origin.

Editorial extensions

If this is right

Outside the origin, the non-characteristic blow-up set is open and the blow-up curve $T$ is of class $C^1$ there, with $|T'(x)|<1/\sqrt{a(x)}$.
The explicit profile in Corollary 3 gives the leading-order shape of $u$ inside the backward cone: a universal power-law singularity whose coefficient depends only on $p$, $\theta(x_0)$, $a(x_0)T'(x_0)^2$, and the optical distance $\varphi(x)-\varphi(x_0)$.
Every characteristic point outside the origin is isolated, and the nearby behavior of $T$ and $T'$ is governed by logarithmic factors with exponent $-((k(x_0)-1)(p-1))/2$ and a constant $\nu(p)$ independent of $\beta$.
At the origin, solutions with non-characteristic data remain bounded in similarity variables in the weighted spaces, showing that the degeneracy does not by itself destroy the scaling framework.
Solutions at characteristic points converge to a sum of $k(x_0)\ge 2$ alternating solitons with centers whose spacing is prescribed by an explicit ODE system inherited from the constant-coefficient theory.

Reading between the lines

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

Editorial inference: the same reduction should make the origin tractable by imposing the radial weight $\rho_0(y)=(1-y^2)^{2/(p-1)-(d-1)/2}y^{d-1}$; Theorem 5 supplies the boundedness input needed to run the standard compactness-plus-Lyapunov argument, so a full origin classification may be within reach if a spectral gap holds in $H^1_{\rho_0}\times L^2_{\rho_0}$.
Editorial inference: if the claimed transfer is correct, the constants in the characteristic asymptotics, in particular $\nu(p)$ and the ODE spacing constant $c_1(p)$, should be independent of $\beta$; a numerical or asymptotic computation of a characteristic blow-up for a strongly varying $\beta$ would test this directly.
Editorial inference: the profile's dependence on $a$ enters only through the optical coordinate $\varphi$ and the product $T'(x_0)\sqrt{a(x_0)}$, suggesting that two media with different $a$ but the same $\varphi$-geometry produce identical normalized blow-up profiles; this is a testable prediction not explicitly stated in the paper.

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

Dear colleague,

You should know up front that the paper's central classification result is off by a constant factor. The authors study a semilinear wave equation with nonconstant degenerate coefficient a(x) and a coefficient b(x) in front of the nonlinear term. They use the optical-length change of variables X = phi(x), which is a sensible move, and derive a similarity-variable equation whose leading nonlinear term is beta(X0)|w|^{p-1}w.

The genuinely new piece is Proposition 2.1, a Lyapunov functional for this equation. The treatment of the new terms I4 and K8, which arise from the nonconstant beta, is careful and appears correct. The radial formulation at the origin, with the weight rho0 in (1.30), is also a reasonable way to address the degeneracy. So there is real mathematical work here.

The problem is what the authors do with that Lyapunov functional. A stationary profile for an equation with coefficient beta(X0) is c times the standard soliton kappa, where c = beta(X0)^{-1/(p-1)}. This follows directly from Lkappa + kappa^p = 0: plugging c kappa into Lw + beta |w|^{p-1}w = 0 gives c^{p-1} = 1/beta. Yet Theorem 2(i), formula (2.29), and Corollary 3 all use the standard kappa with no beta factor. That is not a missing technical detail; it is a scaling error in the central claim. The same error propagates to the characteristic-point statement (1.25) and to the rates in Proposition 4, since those are built from the same solitons.

I should add that the paper largely defers the proofs of the main theorems to "very minor adaptations" of Hamza-Zaag and Merle-Zaag. That would be defensible if the transfer were routine, but with the beta factor wrong it is not. The Lyapunov computation stands, and the error is likely fixable by rescaling w, but as submitted the announced classification is false.

Who should read this: anyone working on blow-up profiles for wave equations with variable coefficients. The Lyapunov part is worth extracting; the classification needs correction. I would send it to a serious referee, but with the expectation of major revision. A referee familiar with the Merle-Zaag machinery will catch the missing factor immediately.

Referee Report

2 major / 4 minor

Summary. The manuscript studies the radial semilinear wave equation (1.1) with nonconstant coefficient a(x) in front of the spatial second-order term and nonconstant coefficient b(x) in the power nonlinearity. By the change of variables X = φ(x) = ∫ dx/√a(x), the equation is rewritten as (1.12), and the authors introduce the similarity variables (1.18). They prove in detail a Lyapunov-functional estimate (Proposition 2.1) in the presence of the new coefficient β(X) = b(x), and then state classification results: outside the origin, non-characteristic points converge exponentially to a single soliton profile and the blow-up curve is C^1, while characteristic points converge to a sum of alternating solitons; at the origin, only a bound in similarity variables is claimed. The proofs of the classification theorems are mostly sketched, with repeated assertions that the arguments of Hamza-Zaag and Merle-Zaag transfer with minor adaptations.

Significance. If the stated classification were correct, it would extend a substantial body of blow-up-profile results to variable-coefficient and degenerate media; the transformation φ, the weighted spaces, and the radial treatment at the origin are natural and potentially useful. Proposition 2.1 is a concrete contribution, since it treats the nonconstant nonlinearity coefficient β(X) with full details, including the new terms I4 and K8. However, the central asymptotic formulas contain an internal scaling inconsistency, and the main theorems are asserted rather than proved, so the significance is presently conditional on substantial corrections.

major comments (2)

[Section 2.2, Theorem 2, Corollary 3] The asymptotic profiles stated in Theorem 2(i)-(ii), formula (2.29), and Corollary 3 are missing the factor b(x0)^{-1/(p-1)}. For a(x) ≡ 1, f = g ≡ 0 and b(x) ≡ β0, the substitution v = β0^{1/(p-1)}u transforms (1.1) into the standard semilinear wave equation, so the blow-up profile of u must be β0^{-1/(p-1)} times the standard κ profile. In self-similar variables, the stationary equation associated with (1.19) is Lw + β(X0)|w|^{p-1}w = 0; since κ solves Lκ + κ^p = 0, the correct stationary profile is cκ with c^{p-1} = β(X0)^{-1}. Formulas (1.24), (1.25), and (2.29) use κ without this factor, so they are false as stated for b(x0) ≠ 1. The energy limit in Theorem 2(ii) should likewise be k(x0)β(x0)^{-2/(p-1)}E(κ0, 0), not k(x0)E(κ0, 0).
[Sections 2.2 and 2.3] The central classification results are not proved in the manuscript. Section 2.2 states that the details will not be given and that the strategy of Hamza-Zaag [6] holds with very minor adaptations, while Section 2.3 justifies Theorem 5 by saying that one can adapt the proof of [6] with no difficulties. In particular, the lower bound in Theorem 1(i), the convergence statements and rates in Theorem 2, Corollary 3, and the characteristic-point asymptotics and isolatedness in Proposition 4 are asserted rather than demonstrated. The load-bearing steps that must be transferred to (1.12) with β(X) ≠ 1 — the covering argument of [9], the spectral gap or mode analysis, and the isolatedness proof of [14] — are not verified in the paper. Since Theorems 2, Corollary 3, and Proposition 4 are the paper's principal claims, this absence of proof is a substantive gap, not a presentational choice.

minor comments (4)

[Corollary 3] The statement begins with the incomplete condition "If R∩R*_+"; it should read "If x0 ∈ R∩R*_+", since R is a set rather than a point.
[Section 2.1, equations (2.11), (2.12), (2.17), (2.23)] The proof of Proposition 2.1 uses N−1 in several displayed estimates, while the corresponding equation (2.3) has d−1 after the change of variables; this notational inconsistency should be corrected.
[Proposition 2.1(i)] The statement quantifies over all X0 > 0, but the admissible range of s depends on X0 through the conditions s ≥ −4 log X0 and s ≥ −log(X0/2); this dependence should be made explicit in the statement.
[Abstract and Theorem 5] The abstract promises "the blow-up behavior and the regularity of the blow-up set," but at the origin Theorem 5 only provides an energy bound in similarity variables; the word "partial" should appear in the abstract as well, or the statements should be aligned with the results actually proved.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the central asymptotic classification is imported from independent prior theorems and a new Lyapunov computation, not reduced by construction to fitted inputs or self-referential definitions.

full rationale

The derivation chain in this paper consists of (1) a change of variables (1.11) that reduces the variable coefficient a(x) to the constant-coefficient form, (2) an explicit Lyapunov functional computation in Section 2.1 that handles the new nonconstant coefficient beta(X), and (3) an assertion that the Merle-Zaag and Hamza-Zaag strategies apply with 'very minor adaptations' or 'with no difficulties.' None of these steps is circular. No parameter is fitted, no data are used, and the claimed blow-up profiles are not defined in terms of the quantities they purport to predict. The main theorems are not proved in full detail and rely heavily on prior work, including the second author's own papers, but those cited results are peer-reviewed, parameter-free theorems with stated assumptions that do not include the target case beta(X) not identically 1; citing them as templates for an adaptation is a rigor and correctness concern, not an equivalence-by-construction. In particular, the apparent omission of a b(x0)^{-1/(p-1)} scaling factor in the asserted profile would be a mathematical error in the transfer, not a circular step. Therefore no circularity is identified.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

No constants in the paper are fitted to data. The integers p and d enter as hypotheses; the constants κ0, c1(p), ν(p) and the shift sequence ᾱi(p,k) are derived in earlier cited work and are not free in this paper. The 'free' element that the paper itself chooses is the class of coefficient functions a, b, f, g satisfying (1.2)-(1.6), which is the domain of the theorems. No new physical or mathematical entities are postulated beyond the standard similarity variables and cones from the cited literature.

free parameters (2)

p
Exponent of the nonlinearity in (1.1); constrained by subcriticality (1.5): 1 < p < (d+3)/(d-1). Exogenous input, not tuned by the authors.
d
Effective dimension fixed by (1.2)-(1.3) via a(x) and N; e.g., d = 2(N-α)/(2-α) for a(x)=|x|^α in (1.7). Exogenous problem parameter, not fitted.

assumptions (4)

domain assumption Structural assumptions (1.2)-(1.6) on a, b, f, g: positivity, finite travel time ∫_0^1 dx/√a(x) < ∞, bounded drift term, subcritical p, subcritical perturbative f, g.
These define the class of equations for which all theorems are stated; they are the contract of the paper, not proved.
domain assumption Global existence and finite-time blow-up with a 1-Lipschitz blow-up time T(x) is assumed to follow from Georgiev-Todorova [16] and Hamza-Zaag [6].
Invoked in Sections 1 and 2 to define T(x) and the blow-up curve Γ; the paper uses it as a black box.
standard math Hardy-Sobolev inequality of Merle-Zaag [8, Appendix B] embeds H^1_ρ into L^{p+1}_ρ.
Used in (1.21) to make the Lyapunov functional well-defined.
ad hoc to paper The entire classification and compactness machinery of Merle-Zaag [8]-[14] and Hamza-Zaag [6] (covering argument, characteristic-point asymptotics, isolatedness) transfers to equation (1.12) with β(X) ≠ 1.
This is the load-bearing transfer assumption; the paper asserts it with 'very minor adaptations' (Section 2.2) and 'with no difficulties' (Section 2.3) rather than proving it.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Classification of the blow-up behavior for a semilinear wave equation with nonconstant degenerate coefficients." pith.science (2026). https://pith.science/paper/F6Q4CRZF

@misc{pith2026190802081,
  author       = {Pith},
  title        = {Pith review of: Classification of the blow-up behavior for a semilinear wave equation with nonconstant degenerate coefficients},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/F6Q4CRZF}},
  note         = {Machine review of arXiv:1908.02081}
}

read the original abstract

We consider a nonlinear wave equation with nonconstant coefficients. In particular, the coefficient in front of the second order space derivative is degenerate. We give the blow-up behavior and the regularity of the blow-up set. Partial results are given at the origin, where the degeneracy occurs. Some nontrivial obstacles, due to the nonconstant speed of propagation, have to be surmounted.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

16 extracted references · 16 canonical work pages

[6]

M. A. Hamza and H. Zaag. Blow-up behavior for the Klein-Go rdon and other perturbed semilinear wave equations. Bull. Sci. Math., 137(8):1087–1109, 2013

work page 2013
[9]

Merle and H

F. Merle and H. Zaag. On growth rate near the blowup surfac e for semilinear wave equations. Int. Math. Res. Not., (19):1127–1155, 2005

work page 2005
[14]

Merle and H

F. Merle and H. Zaag. Isolatedness of characteristic po ints at blowup for a 1-dimensional semilinear wave equation. Duke Math. J, 161(15):2837–2908, 2012

work page 2012
[1]

Alexakis and A

S. Alexakis and A. Shao. On the proﬁle of energy concentra tion at blow-up points for subconformal focusing nonlinear waves. Trans. Amer. Math. Soc., 369(8):5525–5542, 2017

work page 2017
[2]

Antonini and F

C. Antonini and F. Merle. Optimal bounds on positive blow -up solutions for a semilinear wave equation. Internat. Math. Res. Notices, (21):1141–1167, 2001

work page 2001
[3]

Cˆ ote and H

R. Cˆ ote and H. Zaag. Construction of a multisoliton blow up solution to the semilinear wave equation in one space dimension. Comm. Pure Appl. Math., 66(10):1541–1581, 2013

work page 2013
[4]

M. A. Hamza and H. Zaag. Lyapunov functional and blow-up r esults for a class of perturbations of semilinear wave equations in the critical case. J. Hyperbolic Diﬀer. Equ., 9(2):195–221, 2012

work page 2012
[5]

M. A. Hamza and H. Zaag. A Lyapunov functional and blow-up results for a class of perturbed semilinear wave equations. Nonlinearity, 25(9):2759–2773, 2012

work page 2012

Show all 16 references

[7]

M. A. Hamza and H. Zaag. Prescribing the center of mass of a multi-soliton solution for a perturbed semilinear wave equation. J. Diﬀerential Equations, 267(6):3524–3560, 2019

2019
[8]

Merle and H

F. Merle and H. Zaag. Determination of the blow-up rate fo r the semilinear wave equa- tion. Amer. J. Math., 125(5):1147–1164, 2003

2003
[10]

Merle and H

F. Merle and H. Zaag. Existence and universality of the b low-up proﬁle for the semilinear wave equation in one space dimension. J. Funct. Anal., 253(1):43–121, 2007

2007
[11]

Merle and H

F. Merle and H. Zaag. Openness of the set of non-characte ristic points and regularity of the blow-up curve for the 1 D semilinear wave equation. Comm. Math. Phys., 282(1):55– 86, 2008

2008
[12]

Merle and H

F. Merle and H. Zaag. Blow-up behavior outside the origi n for a semilinear wave equation in the radial case. Bull. Sci. Math., 135(4):353–373, 2011

2011
[13]

Merle and H

F. Merle and H. Zaag. Existence and classiﬁcation of cha racteristic points at blow-up for a semilinear wave equation in one space dimension. Amer. J. Math., 134(3):581–648, 2012

2012
[15]

Todorova P

G. Todorova P. Radu and B. Yordanov. Decay estimates for wave equations with variable coeﬃcients. Trans. Amer. Math. Soc., 362(5):2279–2299, 2010. 19

2010
[16]

Georgiev G

V. Georgiev G. Todorova. Existence of a solution of the w ave equation with nonlinear damping and source terms. Journal of Diﬀerential Equations, 109(2):295–308, 1994. Address: Higher Institute for Preparatory Studies in Biology-Geology (ISEP -BG), University of Carthage, 6 Av...

1994

Pith tools

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

[1] [6]

M. A. Hamza and H. Zaag. Blow-up behavior for the Klein-Go rdon and other perturbed semilinear wave equations. Bull. Sci. Math., 137(8):1087–1109, 2013

work page 2013

[2] [9]

Merle and H

F. Merle and H. Zaag. On growth rate near the blowup surfac e for semilinear wave equations. Int. Math. Res. Not., (19):1127–1155, 2005

work page 2005

[3] [14]

Merle and H

F. Merle and H. Zaag. Isolatedness of characteristic po ints at blowup for a 1-dimensional semilinear wave equation. Duke Math. J, 161(15):2837–2908, 2012

work page 2012

[4] [1]

Alexakis and A

S. Alexakis and A. Shao. On the proﬁle of energy concentra tion at blow-up points for subconformal focusing nonlinear waves. Trans. Amer. Math. Soc., 369(8):5525–5542, 2017

work page 2017

[5] [2]

Antonini and F

C. Antonini and F. Merle. Optimal bounds on positive blow -up solutions for a semilinear wave equation. Internat. Math. Res. Notices, (21):1141–1167, 2001

work page 2001

[6] [3]

Cˆ ote and H

R. Cˆ ote and H. Zaag. Construction of a multisoliton blow up solution to the semilinear wave equation in one space dimension. Comm. Pure Appl. Math., 66(10):1541–1581, 2013

work page 2013

[7] [4]

M. A. Hamza and H. Zaag. Lyapunov functional and blow-up r esults for a class of perturbations of semilinear wave equations in the critical case. J. Hyperbolic Diﬀer. Equ., 9(2):195–221, 2012

work page 2012

[8] [5]

M. A. Hamza and H. Zaag. A Lyapunov functional and blow-up results for a class of perturbed semilinear wave equations. Nonlinearity, 25(9):2759–2773, 2012

work page 2012