Convergence of linear solutions through convergence of periodic initial data

Harrison Gaebler; Wesley R Perkins

REVIEW 3 major objections 4 minor 17 references

Convergence of linear solutions through convergence of periodic initial data

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Subharmonic perturbations converge to localized perturbations at the linear level when their initial data converge.

desk verdict Plausible and well-motivated linear-convergence theorem, but the main proof has a load-bearing gap in the uniform Bloch-transform bound, and Corollary 1.4 only proves boundedness, not convergence. read the letter →

arxiv 2505.21762 v1 pith:O4BVYNU5 submitted 2025-05-27 math.AP

Harrison Gaebler , Wesley R Perkins This is my paper

classification math.AP MSC 35B1035B3547D06

keywords subharmonicperturbationslocalizedBlochtransformnormconvergenceoveraperiodlinearevolutionequationssemigroupperiodicwavestabilityRiemannsumapproximation

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 a general linear statement behind a routine formal maneuver in the stability theory of periodic waves: when one approximates a localized perturbation by n-periodic ('subharmonic') perturbations and lets n grow, the corresponding linear evolutions converge. The precise form is Theorem 1.7: if the periodic initial data converge in $L^2(\mathbb{R})$ over a period to a localized datum $g$, the associated extended solutions converge in $L^2(\mathbb{R})$ to the solution with initial data $g$, uniformly in time when the governing semigroup is bounded. The proof works for any linear operator satisfying Assumption 1.5, meaning the same operator generates semigroups on the periodic and line spaces and the Floquet-Bloch fibered operators agree. The result gives a rigorous bridge between two previously separate stability theories: subharmonic (large-but-finite period) and localized (infinite period) perturbations.

What carries the argument

The carrying object is the Floquet–Bloch transform on both the discrete periods $L^2_{\mathrm{per}}(0,nT)$ and the line $L^2(\mathbb{R})$. For an $n$-periodic function the decomposition is a finite sum over the equally spaced quasimomenta $\Omega_n = \{\xi \in [-\pi/T,\pi/T) : e^{i\xi nT}=1\}$, while for a localized function it is the integral over the full Brillouin zone. The paper's key identity is Lemma 2.1, which equates the Bloch transform of the periodic solution with the Bloch transform of the auxiliary line-evolved solution at the discrete quasimomenta; this turns the difference between the two solutions into a Riemann-sum error, which the three preparatory lemmas (pointwise a.e. convergence of Bloch transforms under $L^2$ convergence, and $\xi$-continuity of the Bloch transform on $L^1 \cap H^s$, $s>2$) allow to be estimated term by term.

What would settle it

Look for a sequence $g_n \in L^1_{\mathrm{per}}(0,nT) \cap H^s_{\mathrm{per}}(0,nT)$ that converges in $L^2(\mathbb{R})$ over a period to $g$, whose extended solutions $\tilde v_n(t)$ are dominated by an $L^2$ function for each $t$, but for which $\sup_n |B(w_n)(\xi,x)| = \infty$ on a set of positive $\xi$-measure for some fixed $x$; if the conclusion of Theorem 1.7 fails for this sequence, the theorem is false as stated.

Watch

Extended reading notes

Core claim

The central discovery is that norm convergence of the subharmonic initial data over a period is inherited by the linear solutions, under a mild domination condition on the solutions. Formally, if $g_n \in L^1_{\mathrm{per}}(0,nT) \cap H^s_{\mathrm{per}}(0,nT)$ converge in $L^2(\mathbb{R})$-norm over a period to $g \in L^2(\mathbb{R})$, and the extended solutions $\tilde v_n(t)$ are dominated by an $L^2$ function for each fixed $t$, then $\tilde v_n(t) \to v(t)$ in $L^2(\mathbb{R})$ for each $t \ge 0$. The mechanism is a Riemann-sum comparison: the Bloch decomposition of the $n$-periodic problem is a discrete Riemann sum that approximates the integral Bloch representation of the line problem, and the hypotheses push the difference to zero. If the semigroup $\{e^{tA}\}_{t\ge0}$ on $L^2(\mathbb{R})$ is bounded, the convergence is uniform in time. The paper also records a weak-convergence corollary obtained by applying Banach-Saks averaging to weakly convergent initial data, so that Cesàro means of the periodic data do converge strongly.

Load-bearing premise

The argument requires that, for each fixed point $x$, the Bloch transforms of the auxiliary solutions stay uniformly bounded as $n$ grows; the stated assumptions do not force that bound.

Editorial extensions

If this is right

The formal limit $n\to\infty$ from subharmonic to localized stability results, previously justified only by heuristics, now holds at the linearized level for a general class of operators.
When the linearized semigroup is bounded on $L^2(\mathbb{R})$, the convergence is uniform in time, so long-time estimates pass to the limit.
The result applies to operators arising in Lugiato-Lefever, reaction–diffusion, and KdV/Kuramoto-Sivashinsky stability problems (those satisfying Assumption 1.5), making their subharmonic-to-localized linear comparisons rigorous.
Weak convergence of initial data does not directly give weak convergence of solutions, but Banach–Saks averaging produces a subsequence whose Cesàro means converge strongly over a period (Corollary 4.1).

Reading between the lines

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

The domination hypothesis (iv) is not shown to be implied by the other assumptions; it may itself be a nontrivial a priori bound, and a natural test is whether it holds automatically for operators whose semigroup is bounded on $L^1 \cap H^s$ and $L^2$.
The proof's reliance on a pointwise-in-$\xi$ uniform bound for the Bloch transforms suggests the theorem might extend to settings where only an averaged bound holds, possibly by replacing Egorov's theorem with a more quantitative estimate.
For nonlinear equations, the linear convergence here could serve as a first step to a nonlinear transfer principle: if the nonlinear problem satisfies a suitable continuity estimate in the data, the same subharmonic-to-localized convergence would follow by a bootstrap.
A concrete testable refinement is whether Theorem 1.7 remains true when domination of $\tilde v_n$ is replaced by domination of the initial data only, or when $L^2$ convergence is weakened to $L^p$ convergence with $p \neq 2$.

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.

Desk Editor's Note

read the letter

Short version: the paper has a good idea and the central theorem is plausible, but the proof has a real hole, and the proof of Corollary 1.4 doesn't prove what it claims. I wouldn't accept it as is.

The genuinely new thing here is the attempt to justify, at the linear level, the common practice of recovering localized stability results from the n→∞ limit of subharmonic ones. The Riemann-sum comparison between the discrete Bloch sum over Ω_n and the continuous Bloch integral is the right tool for that, and the Banach-Saks corollary in Section 4 is a clever twist. The motivation from Lugiato-Lefever and reaction-diffusion is concrete, and the self-citations are used for context only, not as load-bearing inputs.

The soft spots are serious. In the proof of Theorem 1.7, the text asserts an M(x) such that |B(w_n)(ξ,x)| ≤ M for all large n, based on pointwise a.e. convergence of B(w_n) to B(v) and the fact that B(v) is in L∞. That simply does not follow. Pointwise convergence doesn't give uniform boundedness, and the L2 domination assumption on \tilde v_n and w_n controls spatial functions, not individual Bloch coefficients. This bound is used to control the endpoint terms and the exceptional set, so without it the whole Riemann-sum estimate collapses. This is a load-bearing gap, not a detail.

Corollary 1.4 also has an issue. The proof shows that \|\tilde v_n(t)-v(t)\| is bounded by a constant Cε, which is a fixed number, not something that goes to zero. That demonstrates boundedness, not strong convergence over a period. The conclusion doesn't follow from the displayed estimate.

There's also a more foundational problem that I think the authors haven't noticed: Definition 1.1's claim that \|g\|_{W^{k,p}(0,nT)} = \|\tilde g\|_{W^{k,p}(R)} is false for k≥1. Zero extension of a periodic function has jumps at ±nT/2, so the extended function is not in W^{k,p}(R). This matters because the whole proof uses H^s norms on R for \tilde g_n, and they may not exist. The proof of Lemma 2.3 also appears to require H^{s+1}, not H^s, though that might be patchable.

The citation pattern is fine. The result may well be repairable by adding uniform L1∩H^s bounds on the initial data (or on \tilde g_n), or an explicit bound on the Bloch transforms, but those assumptions aren't in the paper.

If this lands on my desk, I'd send it to a referee, because the question is relevant and the strategy is sound in concept. But my own recommendation would be reject in current form; I'd invite resubmission after the uniform-bound gap and the corollary are fixed.

Referee Report

3 major / 4 minor

Summary. The paper studies linear initial value problems in which subharmonic initial data g_n ∈ L^1_per(0,nT) ∩ H^s_per(0,nT) converge in L^2(R) over a period to a localized datum g ∈ L^2(R). The main theorem, Theorem 1.7, claims that the corresponding extended solutions \tilde v_n(t) converge to v(t) in L^2(R) for each t ≥ 0, uniformly in t if the semigroup is bounded, under an additional L^2-domination hypothesis on \tilde v_n. The proof uses Floquet–Bloch theory to compare the Riemann-sum representation of the subharmonic solution with the integral representation of the localized solution. The paper also states a motivational corollary for the Lugiato–Lefever equation and a weak-convergence corollary based on the Banach–Saks theorem.

Significance. If Theorem 1.7 were correct, it would provide a general linear justification for the formal limit connecting subharmonic and localized perturbation theories of periodic waves, with potential applications to the stability analyses cited in the introduction. The abstract operator framework of Assumption 1.5 is natural, and the claim is falsifiable through explicit examples. However, the central proof has a load-bearing gap, and the motivating Corollary 1.4 is not proved as written. The paper does not include machine-checked proofs or numerical verification, so the assessment rests entirely on the mathematical arguments.

major comments (3)

[Section 3, proof of Theorem 1.7] The proof asserts, immediately after invoking Lemma 2.2, that from pointwise a.e. convergence of B(w_n)(ξ,x) to B(v)(ξ,x) and the membership B(v)(ξ,x) ∈ L∞_ξ it follows that there exists M = M(x) with |B(w_n)(ξ,x)| ≤ M for a.e. ξ and all but finitely many n. This inference is invalid: pointwise convergence of a sequence of functions to a bounded limit does not imply uniform boundedness in n. The constant M is load-bearing, since it is used to control the endpoint terms via (Δξ + |ξ1+π/T| + |π/T − ξn|) < η/M and the exceptional-set terms via 16M dπ/(nT) < η. Without such a uniform bound, the Riemann-sum comparison has no quantitative control over the difference between subharmonic and extended solutions. Moreover, the auxiliary claim B(v)(ξ,x) ∈ L∞_ξ does not follow from (2.2): for v ∈ L2(R), the Bloch transform is defined as an L2-valued function, and the series representation does not imply a pointwise ξ-uniform bound. Thus Theorem 1.7 is not proved as written.
[Section 1.2, Corollary 1.4] The proof of Corollary 1.4 concludes that the solutions (u_n(t))_n strongly converge over a period to u(t), but the displayed estimate only gives ||\tilde v_n(t) − v(t)||_{L1∩H4} ≤ Cε for a fixed threshold ε = min{ε1, ε2}. The right-hand side is independent of n and does not tend to zero as n → ∞; the argument therefore establishes eventual boundedness, not convergence. Consequently, the advertised "novel corollary" is not proven.
[Section 2.2, Lemma 2.2] The proof of Lemma 2.2 interchanges the limit r → ∞ with the infinite sum over l using the dominated convergence theorem, but the domination required is a sequence a_l ∈ ℓ1 with |\hat f_{n_qr}(ξ + 2πl/T)| ≤ a_l for all r and a.e. ξ. The L2-domination by h2 obtained from Plancherel does not yield such an ℓ1 majorant, since L2 functions need not be summable over the lattice (2π/T)ℤ. As Lemma 2.2 is the source of the pointwise convergence of Bloch transforms used in Theorem 1.7, this gap independently undermines the main proof.

minor comments (4)

[Section 1.2] There is a typo in "Lugiatio-Lefever"; it should be "Lugiato-Lefever".
[Section 1.2, Corollary 1.4 proof] The sentence "By utilizing [8], Theorem 1.3, and ε2 is defined as in [9], Theorem 1.4" is grammatically garbled, and the constants C1 and C2 appear to be swapped between the localized and subharmonic bounds.
[Section 4] In the statement of Corollary 4.1 and its proof, there are typographical errors: "}etA}" should be "{e^{tA}}", and "eGper" should be "\tilde G^{per}" (or similar).
[Section 2.2, Lemma 2.3] In the proof of Lemma 2.3, the bound is stated for s>2, but the Weierstrass M-test only requires s>1; this is not an error but could be clarified.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: Theorem 1.7 is proved from explicit hypotheses with a self-contained Bloch-transform argument; prior self-citations are motivational, not load-bearing.

full rationale

The paper's central result, Theorem 1.7, is derived directly from Assumption 1.5 and the stated L2 convergence and domination hypotheses via the Bloch transform, Lemmas 2.1–2.3, and a Riemann-sum comparison; none of these inputs contains the conclusion. The convergence 'over a period' in the conclusion is not assumed in the hypotheses beyond the analogous convergence of the initial data, and the semigroup is not fitted to target data. Self-citations to [7]–[11] appear in the motivating example, Corollary 1.4, and as examples of operators satisfying Assumption 1.5, but the proof of Theorem 1.7 does not import any theorem from those works as an unverified premise; they are published, externally checkable results rather than circular inputs. The reader's flagged issue—that pointwise convergence of B(w_n) does not imply a uniform-in-n bound M(x) in the proof of Theorem 1.7—is a potential correctness gap in the written estimate, not a circularity, because it does not assume the convergence being proved. The paper's own limitation discussion in Section 4 also honestly notes the difficulty of extending the argument to weak convergence, which is a limitation rather than a circular step. Accordingly, no circular step is present.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The paper introduces no new entities or fitted constants. Its central claim rests on the stated semigroup compatibility assumption, an explicit domination hypothesis, and standard analytic tools. The main weakness is not the axioms themselves but the invalid inference from them inside the proof.

assumptions (3)

domain assumption Assumption 1.5: the Bloch fibers of A_n and A agree on the discrete set Ω_n, i.e. e^{tA_{n,ξ}} = e^{tA_ξ} for each ξ ∈ Ω_n.
This is required for Lemma 2.1 to identify BT(v_n) with B(w_n). It is plausible for periodic-coefficient differential operators but is not automatic for arbitrary linear operators.
ad hoc to paper Theorem 1.7(iv): for each t ≥ 0, the sequence \tilde v_n(·,t) is dominated by a function h_t ∈ L^2(R).
This is imposed to apply the dominated convergence theorem. It is stronger than mere L^2 convergence of the initial data and is not derived from the other assumptions.
standard math Standard Bloch transform, semigroup, Sobolev embedding, Egorov, and Banach-Saks facts are used without proof.
These background tools are standard in the field and are used throughout Sections 2 and 3.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Convergence of linear solutions through convergence of periodic initial data." pith.science (2026). https://pith.science/paper/O4BVYNU5

@misc{pith2026250521762,
  author       = {Pith},
  title        = {Pith review of: Convergence of linear solutions through convergence of periodic initial data},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/O4BVYNU5}},
  note         = {Machine review of arXiv:2505.21762}
}

abstract

When studying the stability of $T$-periodic solutions to partial differential equations, it is common to encounter subharmonic perturbations, i.e. perturbations which have a period that is an integer multiple (say $n$) of the background wave, and localized perturbations, i.e. perturbations that are integrable on the line. Formally, we expect solutions subjected to subharmonic perturbations to converge to solutions subjected to localized perturbations as $n$ tends to infinity since larger $n$ values force the subharmonic perturbation to become more localized. In this paper, we study the convergence of solutions to linear initial value problems when subjected to subharmonic and localized perturbations. In particular, we prove the formal intuition outlined above; namely, we prove that if the subharmonic initial data converges to some localized initial datum, then the linear solutions converge.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

17 extracted references · 17 canonical work pages

[1]

Y. K. Chembo, D. Gomila, M. Tlidi, and C. R. Menyuk. Topical issue: theory and applications of the Lugiato-Lefever equation. Eur. Phys. J. D , 71, 2017. 15

work page 2017
[2]

B. de Rijk. Nonlinear stability and asymptotic behavior of periodic wave trains in reac- tion–diffusion systems against Cub-perturbations. Arch. Rational Mech. Anal. , 248(36), 2024

work page 2024
[3]

Delcey and M

L. Delcey and M. Haragus. Instabilities of periodic waves for the Lugiato-Lefever equation. Rev. Roumaine Math. Pures Appl. , 63(4):377–399, 2018

work page 2018
[4]

Delcey and M

L. Delcey and M. Haragus. Periodic waves of the Lugiato-Lefever equation at the onset of Turing instability. Philos. Trans. Roy. Soc. A , 376(2117):20170188, 21, 2018

work page 2018
[5]

C. Godey. A bifurcation analysis for the Lugiato-Lefever equation. Eur. Phys. J. D , 71:131, 2017

work page 2017
[6]

Hakkaev, M

S. Hakkaev, M. Stanislavova, and A. G. Stefanov. On the generation of stable Kerr frequency combs in the Lugiato-Lefever model of periodic optical waveguides. SIAM J. Appl. Math. , 79(2):477–505, 2019

work page 2019
[7]

Haragus, M

M. Haragus, M. A. Johnson, and W. R. Perkins. Linear modulational and subharmonic dy- namics of spectrally stable Lugiato-Lefever periodic waves. Journal of Differential Equations , 280:315–354, 2021

work page 2021
[8]

Haragus, M

M. Haragus, M. A. Johnson, W. R. Perkins, and B. de Rijk. Nonlinear modulational dynam- ics of spectrally stable Lugiato–Lefever periodic waves. Ann. Inst. H. Poincar´ e Anal. Non Lin´ eaire, 40(4):769–802, 2023

work page 2023

Show all 17 references

[9]

Haragus, M

M. Haragus, M. A. Johnson, W. R. Perkins, and B. de Rijk. Nonlinear subharmonic dynamics of spectrally stable Lugiato–Lefever periodic waves. Communications in Mathematical Physics, 405(227), 2024. https://doi.org/10.1007/s00220-024-05104-5

2024 doi
[10]

M. A. Johnson and W. R. Perkins. Subharmonic dynamics of wave trains in reaction–diffusion systems. Physica D: Nonlinear Phenomena , 422:132891, 2021

2021
[11]

M. A. Johnson and W. R. Perkins. Subharmonic dynamics of wave trains in the korteweg- de vries/kuramoto-sivashinsky equation. Studies in Applied Mathematics , 148(3):1274–1302, 2022

2022
[12]

L. A. Lugiato and R. Lefever. Spatial dissipative structures in passive optical systems. Phys. Rev. Lett., 58, 1987

1987
[13]

Mandel and W

R. Mandel and W. Reichel. A priori bounds and global bifurcation results for frequency combs modeled by the Lugiato-Lefever equation. SIAM J. Appl. Math. , 77(1):315–345, 2017

2017
[14]

Miyaji, I

T. Miyaji, I. Ohnishi, and Y. Tsutsumi. Bifurcation analysis to the Lugiato-Lefever equation in one space dimension. Phys. D , 239(23-24):2066–2083, 2010. 16

2010
[15]

Miyaji, I

T. Miyaji, I. Ohnishi, and Y. Tsutsumi. Stability of a stationary solution for the Lugiato- Lefever equation. Tohoku Math. J. (2) , 63(4):651–663, 2011

2011
[16]

Stanislavova and A

M. Stanislavova and A. G. Stefanov. Asymptotic stability for spectrally stable Lugiato-Lefever solitons in periodic waveguides. J. Math. Phys. , 59(10):101502, 12, 2018

2018
[17]

K. Zumbrun. Forward-modulated damping estimates and nonlocalized stability of periodic Lugiato-Lefever wave. Ann. Inst. H. Poincar´ e Anal. Non Lin´ eaire, 41(2):497–510, 2024. 17

2024

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Y. K. Chembo, D. Gomila, M. Tlidi, and C. R. Menyuk. Topical issue: theory and applications of the Lugiato-Lefever equation. Eur. Phys. J. D , 71, 2017. 15

work page 2017

[2] [2]

B. de Rijk. Nonlinear stability and asymptotic behavior of periodic wave trains in reac- tion–diffusion systems against Cub-perturbations. Arch. Rational Mech. Anal. , 248(36), 2024

work page 2024

[3] [3]

Delcey and M

L. Delcey and M. Haragus. Instabilities of periodic waves for the Lugiato-Lefever equation. Rev. Roumaine Math. Pures Appl. , 63(4):377–399, 2018

work page 2018

[4] [4]

Delcey and M

L. Delcey and M. Haragus. Periodic waves of the Lugiato-Lefever equation at the onset of Turing instability. Philos. Trans. Roy. Soc. A , 376(2117):20170188, 21, 2018

work page 2018

[5] [5]

C. Godey. A bifurcation analysis for the Lugiato-Lefever equation. Eur. Phys. J. D , 71:131, 2017

work page 2017

[6] [6]

Hakkaev, M

S. Hakkaev, M. Stanislavova, and A. G. Stefanov. On the generation of stable Kerr frequency combs in the Lugiato-Lefever model of periodic optical waveguides. SIAM J. Appl. Math. , 79(2):477–505, 2019

work page 2019

[7] [7]

Haragus, M

M. Haragus, M. A. Johnson, and W. R. Perkins. Linear modulational and subharmonic dy- namics of spectrally stable Lugiato-Lefever periodic waves. Journal of Differential Equations , 280:315–354, 2021

work page 2021

[8] [8]

Haragus, M

M. Haragus, M. A. Johnson, W. R. Perkins, and B. de Rijk. Nonlinear modulational dynam- ics of spectrally stable Lugiato–Lefever periodic waves. Ann. Inst. H. Poincar´ e Anal. Non Lin´ eaire, 40(4):769–802, 2023

work page 2023