pith. machine review for the scientific record. sign in

arxiv: 2605.11507 · v1 · submitted 2026-05-12 · 🧮 math.NA · cs.NA

Recognition: no theorem link

A splitting scheme for the wave maps equation at low regularity

Fr\'ed\'eric Rousset, Katharina Schratz, Katie Marsden

Pith reviewed 2026-05-13 01:50 UTC · model grok-4.3

classification 🧮 math.NA cs.NA
keywords wave maps equationLie splitting schemelow regularity convergencediscrete Bourgain spacesnull structurenumerical methods for hyperbolic PDEssubcritical data
0
0 comments X

The pith

A filtered Lie splitting scheme for the wave maps equation converges with initial data in H^s for s greater than 3/2.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper shows that a filtered Lie splitting scheme can be made to converge for the wave maps equation even when the initial data has only low regularity in three space dimensions. The proof works by moving the analysis into discrete Bourgain spaces and by arranging the splitting so that the null structure of the nonlinearity survives at the discrete level. A reader would care because standard numerical methods lose stability or accuracy once the data lose smoothness, and wave maps appear in models of geometric flows where low-regularity data are natural. The main technical obstacle is that the null form involves time derivatives, which introduce extra numerical error; the scheme controls this error sufficiently for convergence to hold on all subcritical data.

Core claim

We prove convergence of a filtered Lie splitting scheme for the wave maps equation with low regularity initial data in dimension 3. The convergence analysis is performed in discrete Bourgain spaces. An important difficulty is that the analysis of wave maps at low regularity requires the use of the null structure of the system; this structure thus has to be preserved at the discrete level to get an effective stable low regularity scheme. Since the null structure involves time derivatives, the scheme has to be designed carefully. The presence of time derivatives in the nonlinearity then constitutes the most significant source of numerical error. Nonetheless, we are able to prove convergence of

What carries the argument

The filtered Lie splitting scheme that preserves the null structure of the wave maps nonlinearity inside discrete Bourgain spaces.

If this is right

  • Convergence holds for every subcritical initial datum in H^s with s > d/2.
  • The dominant error term arising from time derivatives inside the nonlinearity remains controlled.
  • Discrete Bourgain spaces suffice for the entire low-regularity stability analysis.
  • Preservation of the null form at the discrete level is both necessary and sufficient for the scheme to inherit the continuous equation's low-regularity properties.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same discrete preservation strategy could be tested on other null-form wave equations to see whether low-regularity convergence extends beyond wave maps.
  • If the scheme remains stable under mesh refinement below the continuous critical index, it would give a practical tool for exploring singular regimes that are currently inaccessible to standard integrators.

Load-bearing premise

The splitting must keep the null structure of the continuous nonlinearity intact at the discrete level even though time derivatives appear in the nonlinearity.

What would settle it

Numerical experiments showing that solutions generated by the scheme diverge in H^s norm for some initial data with s slightly larger than 3/2 would falsify the convergence claim.

Figures

Figures reproduced from arXiv: 2605.11507 by Fr\'ed\'eric Rousset, Katharina Schratz, Katie Marsden.

Figure 1
Figure 1. Figure 1: Numerical simulation of 1D wave maps by filtered Lie splitting. The initial data is u0pxq “ ` 1{ ? 2p´ cos θpxq ` sin θpxq ¨ sin ϕpxqq, ´ sin θpxq ¨ cos ϕpxq, 1{ ? 2pcos θpxq ` sin θpxq ¨ sin ϕpxqq˘ , v0pxq “ p0, 0, 0q for θpxq “ 2e ´x 2 , ϕpxq “ xe´x 2 . We remark that the upper bound s ă 2 coincides with the lower bound until which Strichartz arguments suffice in the wellposedness theory. We therefore ex… view at source ↗
read the original abstract

We prove convergence of a filtered Lie splitting scheme for the wave maps equation with low regularity initial data in dimension 3. The convergence analysis is performed in discrete Bourgain spaces, as has proved fruitful for the low regularity analysis of the equation in the continuous setting. An important difficulty here is that the analysis of wave maps at low regularity requires the use of the null structure of the system, this structure thus has to be preserved at the discrete level to get an effective stable low regularity scheme. Since the null structure involves time derivatives, the scheme has to be designed carefully. The presence of time derivatives in the nonlinearity then constitutes the most significant source of numerical error. Nonetheless, we are able to prove convergence of the scheme for all subcritical initial data in $H^s$, $s>d/2$.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper claims to prove convergence of a filtered Lie splitting scheme for the 3D wave maps equation with subcritical initial data in H^s (s > 3/2), by carrying out the analysis in discrete Bourgain spaces while carefully preserving the null structure of the nonlinearity. The abstract identifies the time derivatives in the nonlinearity as the dominant source of numerical error but asserts that the scheme is designed so that discrete multipliers and space-time Fourier weights still reproduce the continuous null cancellation uniformly in the time step, yielding convergence for all such data.

Significance. If the estimates close, the result would be a meaningful contribution to low-regularity numerical analysis for nonlinear wave equations. It transfers the null-form techniques that have been successful for the continuous wave maps problem into a stable splitting scheme, potentially allowing reliable computation below the energy space where standard methods lose control. The explicit handling of time derivatives within the discrete Bourgain-space framework is technically nontrivial and, if fully verified, strengthens the case for structure-preserving discretizations in dispersive PDEs.

major comments (2)
  1. [Abstract and convergence analysis] The load-bearing step is the claim that the filtered Lie splitting preserves the null structure involving time derivatives sufficiently well for the discrete Bourgain-space estimates to close at the low-regularity threshold. The abstract states that this preservation is achieved and that time derivatives constitute the most significant error source, yet provides only a sketch of how the discrete multipliers reproduce the continuous cancellation uniformly in the time step. Without the explicit bounds on the resulting error terms (presumably in the main convergence section), it is impossible to confirm that the estimates control all subcritical data in H^s, s > 3/2.
  2. [Convergence analysis] The strongest claim is convergence for all subcritical data. This requires that no loss of cancellation occurs after splitting and filtering; any such loss would prevent the Bourgain-space norms from closing at the continuous threshold. The manuscript must therefore supply the precise discrete null-form estimates that replace the continuous ones, including the dependence on the time step.
minor comments (2)
  1. [Introduction] Notation for the discrete Bourgain spaces and the precise definition of the filtered splitting operator should be introduced with explicit formulas early in the paper to aid readability.
  2. [Introduction] A brief comparison table or paragraph contrasting the continuous null-form estimates with their discrete counterparts would clarify the technical novelty.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thorough review and valuable suggestions on our manuscript. We address each major comment below, providing clarifications on the discrete estimates and indicating the revisions made to strengthen the presentation of the null-form preservation.

read point-by-point responses

  1. Referee: [Abstract and convergence analysis] The load-bearing step is the claim that the filtered Lie splitting preserves the null structure involving time derivatives sufficiently well for the discrete Bourgain-space estimates to close at the low-regularity threshold. The abstract states that this preservation is achieved and that time derivatives constitute the most significant error source, yet provides only a sketch of how the discrete multipliers reproduce the continuous cancellation uniformly in the time step. Without the explicit bounds on the resulting error terms (presumably in the main convergence section), it is impossible to confirm that the estimates control all subcritical data in H^s, s > 3/2.

    Authors: We agree that the abstract provides only a high-level overview and that explicit bounds on the error terms are essential for verifying the low-regularity threshold. The body of the manuscript (particularly the analysis in Section 4) derives the discrete multipliers and shows they reproduce the continuous null cancellation, but the time-step dependence was not stated with full quantitative bounds in one place. In the revised version, we have added Lemma 4.4, which supplies the explicit estimate: the discrete null form differs from the continuous one by an error controlled by C Δt^{1/2} in the discrete Bourgain norm, with C independent of the time step and frequency. This bound is then used directly in the convergence proof in Section 5 to close the estimates for all s > 3/2. The abstract has also been updated to reference this quantitative control. revision: yes

  2. Referee: [Convergence analysis] The strongest claim is convergence for all subcritical data. This requires that no loss of cancellation occurs after splitting and filtering; any such loss would prevent the Bourgain-space norms from closing at the continuous threshold. The manuscript must therefore supply the precise discrete null-form estimates that replace the continuous ones, including the dependence on the time step.

    Authors: The manuscript already contains the required discrete null-form estimates in Section 4 (specifically Lemmas 4.3 and 4.5 and Proposition 4.6), which replace the continuous estimates and explicitly track the time-step dependence, showing that the cancellation is preserved up to terms that remain controllable in the discrete spaces as Δt → 0. No loss occurs that would raise the regularity threshold. To address the referee's request for greater clarity, we have added a short subsection (4.7) that directly compares the discrete and continuous null-form bounds side-by-side, including the precise Δt factors. This makes the absence of any structural loss transparent and confirms convergence for every subcritical datum in H^s, s > 3/2. revision: partial

Circularity Check

0 steps flagged

No circularity: convergence proof is self-contained mathematical argument

full rationale

The paper establishes convergence of a filtered Lie splitting scheme for wave maps via estimates in discrete Bourgain spaces. The abstract and description emphasize that the scheme is designed to preserve the null structure of the nonlinearity (including time derivatives), allowing transfer of continuous low-regularity estimates. No equations, parameters, or claims reduce by construction to fitted inputs, self-definitions, or load-bearing self-citations whose validity depends on the present work. The derivation chain consists of independent analytic estimates rather than tautological renaming or prediction of fitted quantities. This is the standard case of a non-circular proof paper.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The proof rests on standard properties of Bourgain spaces, Sobolev embeddings, and the null form cancellation; no new free parameters or invented entities are introduced in the abstract.

axioms (2)
  • domain assumption Properties of discrete Bourgain spaces allow control of the nonlinearity after splitting
    Invoked to close the convergence estimates at low regularity
  • domain assumption The null structure of the wave maps nonlinearity survives the filtering and splitting operations
    Central to preventing loss of derivatives from time derivatives in the nonlinearity

pith-pipeline@v0.9.0 · 5436 in / 1281 out tokens · 40042 ms · 2026-05-13T01:50:13.148372+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

31 extracted references · 31 canonical work pages · 1 internal anchor

  1. [1]

    S¨ oren Bartels,Semi-implicit approximation of wave maps into smooth or convex surfaces, SIAM J. Numer. Anal. 47(2009), no. 5, 3486–3506. MR2576508

    work page 2009

  2. [2]

    S¨ oren Bartels, Xiaobing Feng, and Andreas Prohl,Finite element approximations of wave maps into spheres, SIAM J. Numer. Anal.46(2007/08), no. 1, 61–87. MR2377255

    work page 2007

  3. [3]

    Comp.78(2009), no

    S¨ oren Bartels, Christian Lubich, and Andreas Prohl,Convergent discretization of heat and wave map flows to spheres using approximate discrete Lagrange multipliers, Math. Comp.78(2009), no. 267, 1269–1292. MR2501050 A SPLITTING SCHEME FOR THE WAVE MAPS EQUATION AT LOW REGULARITY 37

    work page 2009

  4. [4]

    An introduction, Grundlehren der Mathematischen Wis- senschaften, vol

    J¨ oran Bergh and J¨ orgen L¨ ofstr¨ om,Interpolation spaces. An introduction, Grundlehren der Mathematischen Wis- senschaften, vol. No. 223, Springer-Verlag, Berlin-New York, 1976. MR482275

    work page 1976

  5. [5]

    Stefano Francesco Burzio,On long time behavior of solutions to nonlinear dispersive equations, EPFL, 2020

    work page 2020

  6. [6]

    Jiachuan Cao, Buyang Li, Yanping Lin, and Fangyan Yao,Numerical approximation of discontinuous solutions of the semilinear wave equation, SIAM J. Numer. Anal.63(2025), no. 1, 214–238. MR4855504

    work page 2025

  7. [7]

    Grillakis,An introduction to the theory of wave maps and related geometric problems, World Scientific Publishing Co

    Dan-Andrei Geba and Manoussos G. Grillakis,An introduction to the theory of wave maps and related geometric problems, World Scientific Publishing Co. Pte. Ltd., Hackensack, NJ, 2017. MR3585834

    work page 2017

  8. [8]

    Martina Hofmanov´ a and Katharina Schratz,An exponential-type integrator for the KdV equation, Numer. Math. 136(2017), no. 4, 1117–1137. MR3671599

    work page 2017

  9. [9]

    Ignat,A splitting method for the nonlinear Schr¨ odinger equation, J

    Liviu I. Ignat,A splitting method for the nonlinear Schr¨ odinger equation, J. Differential Equations250(2011), no. 7, 3022–3046. MR2771254

    work page 2011

  10. [10]

    Lun Ji, Alexander Ostermann, Fr´ ed´ eric Rousset, and Katharina Schratz,Low regularity full error estimates for the cubic nonlinear Schr¨ odinger equation, SIAM J. Numer. Anal.62(2024), no. 5, 2071–2086. MR4793474

    work page 2024

  11. [11]

    ,Low regularity error estimates for the time integration of 2D NLS, IMA J. Numer. Anal.45(2025), no. 4, 2023–2059. MR4939607

    work page 2025

  12. [12]

    Karper and Franziska Weber,A new angular momentum method for computing wave maps into spheres, SIAM J

    Trygve K. Karper and Franziska Weber,A new angular momentum method for computing wave maps into spheres, SIAM J. Numer. Anal.52(2014), no. 4, 2073–2091. MR3249366

    work page 2014

  13. [13]

    Sergiu Klainerman and Matei Machedon,Smoothing estimates for null forms and applications, 1995, pp. 99–133. A celebration of John F. Nash, Jr. MR1381973

    work page 1995

  14. [14]

    Sergiu Klainerman and Sigmund Selberg,Bilinear estimates and applications to nonlinear wave equations, Commun. Contemp. Math.4(2002), no. 2, 223–295. MR1901147

    work page 2002

  15. [15]

    Math.149(2021), no

    Buyang Li and Yifei Wu,A fully discrete low-regularity integrator for the 1D periodic cubic nonlinear Schr¨ odinger equation, Numer. Math.149(2021), no. 1, 151–183. MR4312402

    work page 2021

  16. [16]

    ,An unfiltered low-regularity integrator for the KdV equation with solutions belowh 1, Found. Comput. Math.62(2025). MR4462616

    work page 2025

  17. [17]

    Yongsheng Li, Yifei Wu, and Fangyan Yao,Convergence of an embedded exponential-type low-regularity integrators for the KdV equation without loss of regularity, Ann. Appl. Math.37(2021), no. 1, 1–21. MR4284062

    work page 2021

  18. [18]

    J.72(1993), no

    Hans Lindblad,A sharp counterexample to the local existence of low-regularity solutions to nonlinear wave equations, Duke Math. J.72(1993), no. 2, 503–539. MR1248683

    work page 1993

  19. [19]

    Katie Marsden,Probabilistic and deterministic wellposedness for low regularity dispersive equations, EPFL, 2024

    work page 2024

  20. [20]

    ,Global solutions to the 3D half-wave maps equation with angular regularity, Calc. Var. Partial Differential Equations65(2026), no. 1, Paper No. 32, 82. MR5001438

    work page 2026

  21. [21]

    Comp.91(2021), no

    Alexander Ostermann, Fr´ ed´ eric Rousset, and Katharina Schratz,Error estimates at low regularity of splitting schemes for NLS, Math. Comp.91(2021), no. 333, 169–182. MR4350536

    work page 2021

  22. [22]

    Com- put

    ,Error estimates of a Fourier integrator for the cubic Schr¨ odinger equation at low regularity, Found. Com- put. Math.21(2021), no. 3, 725–765. MR4269650

    work page 2021

  23. [23]

    ,Fourier integrator for periodic NLS: low regularity estimates via discrete Bourgain spaces, J. Eur. Math. Soc. (JEMS)25(2023), no. 10, 3913–3952. MR4634686

    work page 2023

  24. [24]

    Fr´ ed´ eric Rousset and Katharina Schratz,A general framework of low regularity integrators, SIAM J. Numer. Anal. 59(2021), no. 3, 1735–1768. MR4275500

    work page 2021

  25. [25]

    Anal.4(2022), no

    ,Convergence error estimates at low regularity for time discretizations of KdV, Pure Appl. Anal.4(2022), no. 1, 127–152. MR4419370

    work page 2022

  26. [26]

    Maximilian Ruff,Improved error estimates for low-regularity integrators using space-time bounds, arXiv:2503.22621

    work page internal anchor Pith review Pith/arXiv arXiv

  27. [27]

    Maximilian Ruff and Roland Schnaubelt,Error analysis of the Lie splitting for semilinear wave equations with finite-energy solutions, Discrete Contin. Dyn. Syst.45(2025), no. 9, 2969–3008. MR4893133

    work page 2025

  28. [28]

    Thesis (Ph.D.)–Princeton University

    Sigmund Selberg,Multilinear space-time estimates and applications to local existence theory for nonlinear wave equations, ProQuest LLC, Ann Arbor, MI, 1999. Thesis (Ph.D.)–Princeton University. MR2699015

    work page 1999

  29. [29]

    106, Conference Board of the Mathematical Sciences, Washington, DC; by the American Mathematical Society, Providence, RI, 2006

    Terence Tao,Nonlinear dispersive equations, CBMS Regional Conference Series in Mathematics, vol. 106, Conference Board of the Mathematical Sciences, Washington, DC; by the American Mathematical Society, Providence, RI, 2006. Local and global analysis. MR2233925

    work page 2006

  30. [30]

    Comp.91(2022), no

    Yifei Wu and Fangyan Yao,A first-order Fourier integrator for the nonlinear Schr¨ odinger equation onTwithout loss of regularity, Math. Comp.91(2022), no. 335, 1213–1235. MR4405493

    work page 2022

  31. [31]

    3, 1049–1090

    Yifei Wu and Xiaofei Zhao,Embedded exponential-type low-regularity integrators for KdV equation under rough data, BIT62(2022), no. 3, 1049–1090. MR4462616 38 K. MARSDEN, F. ROUSSET, AND K. SCHRATZ Department of Mathematics, University of California, Los Angeles, CA 90095, USA (K.Marsden) Email address:kmarsden@math.ucla.edu Laboratoire de Math´ematiques d...

    work page 2022