pith. sign in

arxiv: 2606.13219 · v1 · pith:3UNMFAQ4new · submitted 2026-06-11 · 🧮 math.NA · cs.NA

Embedded Trefftz DG method for steady Navier-Stokes flow. Part II: Nonlinear problem

Paul Stocker , Igor Voulis , Christoph Lehrenfeld , Philip L. Lederer This is my paper

Pith reviewed 2026-06-27 06:07 UTC · model grok-4.3

classification 🧮 math.NA cs.NA
keywords Trefftz DGNavier-StokesPicard iterationembedded methodOseen problema priori error analysisdiscontinuous Galerkinincompressible flow
0
0 comments X

The pith

Projections between convection-dependent Trefftz spaces allow proofs of existence, uniqueness and Picard convergence for an embedded Trefftz DG discretization of the steady Navier-Stokes equations.

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

The paper shows how to handle the fact that the discrete Trefftz space for the reduced Oseen problem changes with each Picard iterate because it depends on the current convection field. By constructing stable projections between these spaces, the authors control the solution map and establish that discrete solutions exist and are unique under small-data assumptions, that the Picard iteration converges to them, and that the method inherits a priori error bounds from compatible DG schemes. This construction resolves the main technical difficulty in extending the linear analysis of Part I to the nonlinear case. The theory is illustrated by numerical experiments on standard incompressible flow benchmarks.

Core claim

The central claim is that suitable projections between the convection-dependent Trefftz spaces permit an analysis of the nonlinear problem by controlling the reduced Oseen solution map, thereby proving existence of discrete solutions, uniqueness, convergence of the Picard iteration, and an a priori error estimate that inherits the convergence properties of the underlying DG discretization.

What carries the argument

Projections between convection-dependent Trefftz spaces used to control the reduced Oseen solution map.

If this is right

  • Discrete solutions exist and are unique under the stated resolution and small-data assumptions.
  • The Picard iteration converges to the unique discrete solution.
  • An a priori error bound holds by direct comparison with a compatible DG Navier-Stokes discretization.
  • The numerical method performs as predicted on standard benchmark problems for incompressible flow.

Where Pith is reading between the lines

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

  • This projection technique could apply to other discretizations where the finite-element space depends on the current solution iterate.
  • Extending the method to time-dependent or three-dimensional flows would require analogous control of the space changes across time steps.
  • The stability of the projections might be verified computationally on specific meshes to confirm the analysis assumptions hold in practice.

Load-bearing premise

The projections between the convection-dependent Trefftz spaces exist and are stable with bounds independent of the convection field.

What would settle it

Numerical computation showing that the norm of the projection operator between two Trefftz spaces for nearby convection fields grows unboundedly as the mesh is refined would falsify the stability assumption needed for the proofs.

Figures

Figures reproduced from arXiv: 2606.13219 by Christoph Lehrenfeld, Igor Voulis, Paul Stocker, Philip L. Lederer.

Figure 1
Figure 1. Figure 1: Convergence of the velocity and pressure in the [PITH_FULL_IMAGE:figures/full_fig_p019_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Number of Picard iterations needed to reach the tolerance [PITH_FULL_IMAGE:figures/full_fig_p020_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Streamlines of the velocity field for the Schäfer–Turek benchmark problem computed [PITH_FULL_IMAGE:figures/full_fig_p021_3.png] view at source ↗
read the original abstract

We develop and analyze an embedded Trefftz-DG method for the steady incompressible Navier-Stokes equations, based on the reduced Oseen discretization from Part I. The main difficulty is that the reduced Trefftz space depends on the convection field, so successive Picard iterates live in different discrete spaces. We address this by constructing projections between convection-dependent Trefftz spaces and using them to control the reduced Oseen solution map. Under suitable resolution and small-data assumptions, we prove existence of discrete solutions, uniqueness, and convergence of the Picard iteration. We also derive an a priori error analysis by relating the method to the underlying DG discretization, thereby inheriting convergence properties from compatible DG Navier-Stokes analyses. Numerical experiments on standard incompressible-flow benchmarks illustrate the theory.

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 manuscript develops an embedded Trefftz DG method for the steady incompressible Navier-Stokes equations, extending the reduced Oseen discretization from Part I. The central technical step is the construction of projections between convection-dependent Trefftz spaces to accommodate the space dependence arising in the Picard iteration. Under small-data and resolution assumptions, the paper proves existence and uniqueness of discrete solutions, convergence of the Picard iteration, and an a priori error bound obtained by relating the method to an underlying standard DG discretization. Numerical experiments on incompressible-flow benchmarks are included to support the analysis.

Significance. If the projection construction and its stability properties hold with constants independent of the convection field, the work would provide a rigorous route to apply Trefftz DG methods to nonlinear problems while inheriting convergence rates from established DG Navier-Stokes theory. The explicit handling of space dependence via projections and the reduction to a compatible DG analysis are the main strengths; the numerical results on standard benchmarks add concrete validation.

major comments (2)
  1. [§3.3] §3.3 (Projection construction): The maps Π_{k,k+1} : V_h(β^k) → V_h(β^{k+1}) are asserted to be stable in the DG norm and to satisfy an approximation property, yet the proof supplies no explicit bound on the operator norm that is independent of ||β^k||_{L^∞}. This bound is load-bearing for the contraction-mapping argument that closes existence and convergence of the Picard iteration under the stated small-data assumption.
  2. [§4.2, Eq. (4.8)] §4.2, Eq. (4.8): The a priori error analysis reduces the Trefftz solution to the underlying DG solution plus projection errors; however, the approximation property used for the convection-dependent spaces is only stated qualitatively, without a quantitative estimate that would guarantee the same convergence rate as the reference DG method when the convection field varies between iterates.
minor comments (2)
  1. [§2] The notation for the reduced Trefftz spaces V_h(β) is introduced without a compact reference table listing the precise polynomial degree and embedding conditions; adding such a table would improve readability.
  2. [§5] Figure 5.1 caption does not state the mesh size or polynomial degree used for the lid-driven cavity example, making direct comparison with the theory difficult.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments on the manuscript. The two major comments raise important points about the explicit stability and approximation properties of the projections between convection-dependent Trefftz spaces. We address each below and will incorporate the requested clarifications and quantitative estimates into a revised version.

read point-by-point responses

  1. Referee: [§3.3] §3.3 (Projection construction): The maps Π_{k,k+1} : V_h(β^k) → V_h(β^{k+1}) are asserted to be stable in the DG norm and to satisfy an approximation property, yet the proof supplies no explicit bound on the operator norm that is independent of ||β^k||_{L^∞}. This bound is load-bearing for the contraction-mapping argument that closes existence and convergence of the Picard iteration under the stated small-data assumption.

    Authors: We agree that an explicit operator-norm bound independent of ||β^k||_{L^∞} is necessary to close the contraction-mapping argument rigorously. Under the small-data assumption the velocity fields remain uniformly bounded in L^∞ independently of the iteration index, but this control was only used implicitly. In the revision we will insert a new lemma in §3.3 that derives the bound ||Π_{k,k+1} v||_{DG} ≤ C ||v||_{DG} with C depending only on the mesh regularity, polynomial degree, and the small-data constant, thereby making the independence explicit and completing the contraction estimate. revision: yes

  2. Referee: [§4.2, Eq. (4.8)] §4.2, Eq. (4.8): The a priori error analysis reduces the Trefftz solution to the underlying DG solution plus projection errors; however, the approximation property used for the convection-dependent spaces is only stated qualitatively, without a quantitative estimate that would guarantee the same convergence rate as the reference DG method when the convection field varies between iterates.

    Authors: The referee correctly identifies that the current statement of the approximation property is qualitative. Because the projections are constructed from the same underlying polynomial spaces and the convection fields are controlled by the small-data assumption, the projection error admits the same order as the standard DG approximation error. We will add a quantitative lemma (with constants independent of the particular β^k) immediately before Eq. (4.8) and substitute it into the error decomposition; this will confirm that the Trefftz method inherits the optimal rate of the reference DG scheme without degradation. revision: yes

Circularity Check

0 steps flagged

Derivation self-contained; no reduction to inputs by construction

full rationale

The paper explicitly constructs projections between convection-dependent Trefftz spaces to control the reduced Oseen map across Picard iterates and derives a priori error estimates by relating the embedded Trefftz method to an underlying DG discretization whose stability and approximation properties are invoked as external. No equation or claim reduces a main result (existence, uniqueness, Picard convergence) to a fitted parameter, self-definition, or prior self-citation by construction; the Part I reference supplies the linear reduced Oseen scheme but is not load-bearing for the nonlinear analysis under the stated small-data and resolution assumptions. The derivation chain therefore remains independent of its own outputs.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Review performed on abstract only; the paper invokes standard background results from DG theory for Navier-Stokes together with small-data and resolution assumptions whose precise statements are not visible here.

axioms (2)
  • domain assumption Standard mathematical assumptions guaranteeing well-posedness of the continuous Navier-Stokes problem under small data
    Invoked via the phrase 'small-data assumptions' to obtain discrete existence and uniqueness.
  • ad hoc to paper Existence and boundedness of projections between distinct convection-dependent Trefftz spaces
    Constructed explicitly to control the reduced Oseen solution map across Picard iterates.

pith-pipeline@v0.9.1-grok · 5665 in / 1385 out tokens · 26995 ms · 2026-06-27T06:07:38.087853+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

26 extracted references · 16 canonical work pages · 1 internal anchor

  1. [1]

    Cockburn, G

    B. Cockburn, G. Kanschat, and D. Schötzau. A locally conservative LDG method for the incompressible Navier-Stokes equations.Math. Comput., 74(251):1067–1095, 2005.doi: 10.1090/S0025-5718-04-01718-1

    work page doi:10.1090/s0025-5718-04-01718-1 2005

  2. [2]

    D. A. Di Pietro and A. Ern. Discrete functional analysis tools for discontinuous Galerkin methods with application to the incompressible Navier-Stokes equations.Math. Comput., 79(271):1303–1330, 2010.doi:10.1090/S0025-5718-10-02333-1

    work page doi:10.1090/s0025-5718-10-02333-1 2010

  3. [3]

    D. A. Di Pietro and A. Ern.Mathematical aspects of discontinuous Galerkin meth- ods, volume 69. Springer Science & Business Media, Heidelberg, 2011.doi:10.1007/ 978-3-642-22980-0

    2011

  4. [4]

    Gómez, C

    S. Gómez, C. Perinati, and P. Stocker. Inf-sup stable space-time local discontinuous Galerkin method for the heat equation.Journal of Scientific Computing, 106(1):22, 2025.doi: 10.1007/s10915-025-03121-7

    work page doi:10.1007/s10915-025-03121-7 2025

  5. [5]

    C. Heil. Space-time Trefftz DG methods for parabolic PDEs.GRO.data, 2024.doi: 10.25625/ZSA8UU

    work page doi:10.25625/zsa8uu 2024

  6. [6]

    Heimann, C

    F. Heimann, C. Lehrenfeld, P. Stocker, and H. von Wahl. Unfitted Trefftz discontinuous Galerkin methods for elliptic boundary value problems.ESAIM: M2AN, 2023.doi:10. 1051/m2an/2023064

    arXiv 2023

  7. [7]

    Imbert-Gérard, A

    L.-M. Imbert-Gérard, A. Moiola, C. Perinati, and P. Stocker. Polynomial quasi-Trefftz DG for PDEs with smooth coefficients: elliptic problems.IMA Journal of Numerical Analysis, page drae094, 2025.doi:10.1093/imanum/drae094

    work page doi:10.1093/imanum/drae094 2025

  8. [8]

    P. L. Lederer, C. Lehrenfeld, and P. Stocker. Trefftz discontinuous Galerkin discretization for the Stokes problem.Numerische Mathematik, 2024.doi:10.1007/s00211-024-01404-z

    work page doi:10.1007/s00211-024-01404-z 2024

  9. [9]

    Lehrenfeld and P

    C. Lehrenfeld and P. Stocker. Embedded Trefftz discontinuous Galerkin methods.Int. J. Numer. Methods Eng., 124(17):3637–3661, 2023.doi:10.1002/nme.7258

    work page doi:10.1002/nme.7258 2023

  10. [10]

    Lehrenfeld, P

    C. Lehrenfeld, P. Stocker, and M. Zienecker. Sparsity comparison of polytopal finite element methods.PAMM, 24(3):e202400150, 2024.doi:10.1002/pamm.202400150

    work page doi:10.1002/pamm.202400150 2024

  11. [11]

    Lozinski

    A. Lozinski. A primal discontinuous Galerkin method with static condensation on very general meshes.Numer. Math., 143(3):583–604, 2019.doi:10.1007/s00211-019-01067-1

    work page doi:10.1007/s00211-019-01067-1 2019

  12. [12]

    J. C. Meyer, C. Lehrenfeld, and I. Voulis. On the conforming trefftz finite element method and applications.Zenodo, Oct. 2025.doi:10.5281/zenodo.17307511

    work page doi:10.5281/zenodo.17307511 2025

  13. [13]

    Montlaur, S

    A. Montlaur, S. Fernandez-Mendez, and A. Huerta. Discontinuous Galerkin methods for the Stokes equations using divergence-free approximations.International Journal for Numerical Methods in Fluids, 57(9):1071–1092, 2008.doi:10.1002/fld.1716. 22

    work page doi:10.1002/fld.1716 2008

  14. [14]

    Montlaur, S

    A. Montlaur, S. Fernandez-Mendez, J. Peraire, and A. Huerta. Discontinuous Galerkin methods for the Navier–Stokes equations using solenoidal approximations.International Journal for Numerical Methods in Fluids, 64(5):549–564, 2010

    2010

  15. [15]

    Heidelberg: Univ

    G.Nabh.On higher order methods for the stationary incompressible Navier-Stokes equations. Heidelberg: Univ. Heidelberg, Naturwissenschaftlich-Mathematische Gesamtfakultät, 1998

    1998

  16. [16]

    A discontinuous Galerkin method for elliptic-hyperbolic equations

    C. Perinati, L.-M. Imbert-Gérard, A. Moiola, and P. Stocker. A discontinuous Galerkin method for elliptic-hyperbolic equations.arXiv preprint arXiv:2604.06910, 2026. URL: https://arxiv.org/abs/2604.06910,doi:10.48550/arXiv.2604.06910

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2604.06910 2026

  17. [17]

    Schäfer and S

    M. Schäfer and S. Turek. Benchmark computations of laminar flow around a cylinder. (With support by F. Durst, E. Krause and R. Rannacher). InFlow simulation with high- performance computers II. DFG priority research programme results 1993 - 1995, pages 547–566. Wiesbaden: Vieweg, 1996

    1993

  18. [18]

    Schlesinger

    E. Schlesinger. Embedded Trefftz Trace DG Methods for PDEs on unfitted Surfaces. GRO.data, 2023.doi:10.25625/QTOPWD

    work page doi:10.25625/qtopwd 2023

  19. [19]

    Schöberl

    J. Schöberl. C++ 11 implementation of finite elements in NGSolve.Institute for analysis and scientific computing, Vienna University of Technology, 30, 2014

    2014

  20. [20]

    P. Stocker. NGSTrefftz: Add-on to NGSolve for Trefftz methods.J. Open Source Softw., 7(71):4135, 2022.doi:10.21105/joss.04135

    work page doi:10.21105/joss.04135 2022

  21. [21]

    Stocker and I

    P. Stocker and I. Voulis. Embedded Trefftz DG method for the Helmholtz equation.arXiv preprint arXiv:2603.13034, 2026.doi:10.48550/arXiv.2603.13034

    work page doi:10.48550/arxiv.2603.13034 2026

  22. [22]

    Stocker, I

    P. Stocker, I. Voulis, P. L. Lederer, and C. Lehrenfeld. Embedded Trefftz DG method for steady Navier–Stokes flow. Part I: Oseen linearization.arXiv preprint, 2026

    2026

  23. [23]

    https://doi.org/10.5281/zenodo

    P. Stocker, I. Voulis, C. Lehrenfeld, and P. L. Lederer. Replication Data for: Embed- ded Trefftz DG method for steady Navier–Stokes flow, June 2026.doi:10.5281/zenodo. 20490547

    work page doi:10.5281/zenodo 2026

  24. [24]

    E. Trefftz. Ein Gegenstück zum Ritzschen Verfahren.Proc. 2nd Int. Cong. Appl. Mech., Zurich, 1926, pages 131–137, 1926

    1926

  25. [25]

    J. Yang, M. Potier-Ferry, K. Akpama, H. Hu, Y. Koutsawa, H. Tian, and D. S. Zézé. Trefftz methods and Taylor series.Arch. Comput. Methods Eng., 27(3):673–690, 2020

    2020

  26. [26]

    Zeidler.Nonlinear Functional Analysis and its Applications I: Fixed-Point Theorems

    E. Zeidler.Nonlinear Functional Analysis and its Applications I: Fixed-Point Theorems. Springer, New York, 1986. 23

    1986