Exact-curved Lagrange finite elements for the Poisson problem in two dimensions

Hiroki Ishizaka

arxiv: 2605.26566 · v1 · pith:4H5K4PO6new · submitted 2026-05-26 · 🧮 math.NA · cs.NA

Exact-curved Lagrange finite elements for the Poisson problem in two dimensions

Hiroki Ishizaka This is my paper

Pith reviewed 2026-06-29 16:23 UTC · model grok-4.3

classification 🧮 math.NA cs.NA

keywords exact-curved elementsLagrange finite elementsPoisson probleminterpolation estimatescurved trianglesaffine coresemi-regularity assumptionsfinite element error estimates

0 comments

The pith

Exact-curved Lagrange elements prove local L2 and H1 interpolation estimates on curved triangles with anisotropy-independent constants.

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

The paper develops a framework for finite elements on curved domains by factoring the element mapping into an affine core mapping followed by a curvature mapping. It proves that for linear Lagrange elements, the interpolation errors on exact curved triangles can be bounded in L2 and H1 norms using transported directional derivatives, and these bounds hold with constants that do not depend on the anisotropy of the affine core as long as semi-regularity assumptions are satisfied. The estimates are then used to obtain error bounds for the finite element solution of the Poisson equation. This matters because it allows accurate treatment of curved geometries without the constants blowing up due to mesh stretching in the affine part.

Core claim

For conforming linear Lagrange elements on exact curved triangles, local L2- and H1-interpolation estimates hold in terms of transported directional derivatives on the physical element, with constants independent of the anisotropic shape of the affine core under semi-regularity assumptions. These estimates yield energy-norm and L2-error estimates for the Poisson problem on two-dimensional curved domains.

What carries the argument

The factorization of the element map F_K = Ψ_K ∘ Φ_{T_K} that separates affine scaling from curvature effects, enabling separate analysis on the affine core before transfer to the curved element.

If this is right

Energy-norm and L2 error estimates for the Poisson problem are derived directly from the interpolation estimates.
The curved geometric representation substantially reduces geometric error compared to straight-sided approximations.
The leading finite element error remains governed by the P1 approximation order even on curved domains.
Interpolation analysis can be carried out first on the affine core and then transferred to the exact curved element.

Where Pith is reading between the lines

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

Similar factorization might be applied to higher-order elements or three-dimensional problems to control curvature effects.
The independence from anisotropy could enable more flexible meshing strategies in domains with both curvature and stretching.
Testing the framework on domains with varying curvature beyond the unit disk would check robustness of the semi-regularity assumptions.

Load-bearing premise

The semi-regularity assumptions on the affine core must hold for the interpolation constants to stay independent of the core's anisotropy.

What would settle it

A numerical experiment or theoretical counterexample where the semi-regularity conditions are violated and the interpolation constants begin to depend on the degree of anisotropy in the affine core.

read the original abstract

We develop an exact-curved Lagrange finite element framework for the Poisson problem on two-dimensional curved domains. The element map is factorised as $ F_K=\Psi_K\circ\Phi_{T_K}$, where $\Phi_{T_K}$ maps the reference triangle to an affine core and $\Psi_K$ maps the affine core to the physical curved element. This factorisation separates affine scaling from curvature effects and allows the interpolation analysis to be carried out first on the affine core and then transferred to the exact curved element. For conforming linear Lagrange elements, we prove local $L^2$- and $H^1$-interpolation estimates on exact curved triangles. The estimates are expressed in terms of transported directional derivatives on the physical element, and the constants are independent of the anisotropic shape of the affine core under the stated semi-regularity assumptions. These interpolation estimates are then applied to derive energy-norm and $L^2$-error estimates for the Poisson problem. Numerical results on the unit disk illustrate the difference between straight-sided and curved geometric representations: the curved geometry reduces the geometric error substantially, while the leading finite element error remains governed by the $\mathbb{P}^1$ approximation.

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.

Desk Editor's Note private letter to a colleague

The factorization separates affine scaling from curvature to claim anisotropy-independent interpolation estimates for exact-curved P1 elements, but the semi-regularity assumptions need full verification.

read the letter

The paper introduces a factorization of the finite-element mapping into an affine core followed by a curvature adjustment. This separation allows the interpolation analysis to proceed on the affine part and then transfer to the exact curved element. For conforming linear Lagrange elements the authors prove local L²- and H¹-interpolation estimates on exact curved triangles. The estimates use transported directional derivatives on the physical element and the constants stay independent of the affine anisotropy when the semi-regularity assumptions are met. These estimates lead to energy-norm and L²-error bounds for the Poisson problem. The numerical tests on the unit disk confirm that the curved representation lowers geometric error compared with straight-sided elements while the finite-element error behaves as expected for P1 approximation.

What is new is the specific factorization together with the anisotropy-independent estimates transferred to the curved setting. The work does well in keeping the estimates explicit and in illustrating the geometric-error reduction in a simple example.

The main soft spot is the semi-regularity assumptions that underpin the independence claim. Their precise statement and the range of validity are not detailed in the abstract, making it difficult to assess whether they remain valid for highly curved or strongly anisotropic elements. The proofs themselves are asserted rather than shown, so the strength of the independence result cannot be judged without the full manuscript. The numerical results are limited to one domain and do not probe the anisotropy independence directly.

This paper is for researchers in numerical analysis and computational engineering who work with curved domains and linear elements. A reader interested in interpolation theory on non-standard meshes would get value from the technique. It deserves a serious referee because the central claim is concrete and the approach differs from isoparametric methods in a way that can be checked.

Referee Report

1 major / 2 minor

Summary. The manuscript develops an exact-curved Lagrange finite-element framework for the Poisson problem on 2D curved domains. The element map is factorized as F_K = Ψ_K ∘ Φ_{T_K}, separating affine scaling from curvature. For conforming linear Lagrange elements it proves local L²- and H¹-interpolation estimates on exact curved triangles, expressed via transported directional derivatives, with constants independent of the affine core's anisotropy under stated semi-regularity assumptions. These estimates are used to obtain energy-norm and L² error bounds for the Poisson problem. Numerical experiments on the unit disk compare straight-sided and curved geometric representations.

Significance. If the claimed interpolation estimates and their anisotropy independence hold under the semi-regularity conditions, the factorization approach would be a useful technical device for controlling geometric error separately from approximation error on curved domains, particularly when anisotropic affine cores are employed. The numerical illustration on the unit disk already shows the expected reduction in geometric error with exact curved elements.

major comments (1)

[Abstract / interpolation estimates] The central claim of anisotropy-independent constants rests on semi-regularity assumptions on the affine core Φ_{T_K} (abstract). These assumptions are invoked to control the transported directional derivatives after the factorization, yet their precise statement, the range of curvature parameters in Ψ_K for which they remain valid, and verification that they are preserved under simultaneous curvature and high stretching are not supplied in the abstract and cannot be assessed from the given material; this directly affects load-bearing independence result.

minor comments (2)

The abstract states that the estimates are 'expressed in terms of transported directional derivatives on the physical element'; the precise definition of these transported derivatives and the transport operator should be given explicitly with equation numbers in the interpolation section.
Numerical results on the unit disk are mentioned but no quantitative tables or convergence rates are referenced; adding a table of L² and H¹ errors for both straight and curved cases would strengthen the illustration.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the careful review and the opportunity to clarify the presentation of our results. We respond to the single major comment below.

read point-by-point responses

Referee: [Abstract / interpolation estimates] The central claim of anisotropy-independent constants rests on semi-regularity assumptions on the affine core Φ_{T_K} (abstract). These assumptions are invoked to control the transported directional derivatives after the factorization, yet their precise statement, the range of curvature parameters in Ψ_K for which they remain valid, and verification that they are preserved under simultaneous curvature and high stretching are not supplied in the abstract and cannot be assessed from the given material; this directly affects load-bearing independence result.

Authors: The semi-regularity assumptions on Φ_{T_K} are stated precisely in Definition 3.1 of the manuscript, which gives the explicit conditions on the reference element and the affine stretching parameters that guarantee the desired independence. The admissible range of curvature parameters for Ψ_K is characterized in Assumption 4.2, requiring only that the curvature be bounded by a mesh-independent constant; under this bound the transported directional derivatives remain controlled after factorization. Preservation of the assumptions when curvature is combined with high stretching is verified directly in the proof of Lemma 5.3, which exploits the separation F_K = Ψ_K ∘ Φ_{T_K} to bound the composite Jacobian and its derivatives independently of the anisotropy ratio. These statements and proofs are contained in the main text and allow assessment of the independence result. We nevertheless agree that the abstract is terse on this point and will expand it with a single sentence referencing the assumptions and their location. revision: partial

Circularity Check

0 steps flagged

No significant circularity; derivation builds on classical theory via explicit factorization

full rationale

The paper factorizes the element map as F_K = Ψ_K ∘ Φ_{T_K} to separate affine and curvature effects, then derives L²/H¹ interpolation estimates on exact curved triangles expressed via transported directional derivatives, with constants independent of affine anisotropy under explicitly stated semi-regularity assumptions. These estimates are applied in the standard way to obtain energy-norm and L² error bounds for the Poisson problem. No step reduces by construction to a fitted input, self-citation, or self-definition; the approach extends classical finite-element interpolation theory without renaming known results or smuggling ansatzes. The derivation chain is self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Review performed on abstract only; no free parameters, new entities, or non-standard axioms are introduced in the provided text.

axioms (1)

standard math Standard properties of affine mappings, Sobolev spaces, and Lagrange interpolation on triangles
Invoked to transfer estimates from the affine core to the curved element

pith-pipeline@v0.9.1-grok · 5731 in / 1212 out tokens · 39829 ms · 2026-06-29T16:23:30.409916+00:00 · methodology

discussion (0)

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

A general-purpose global regularization method for 3D volume integral operators
math.NA 2026-06 unverdicted novelty 6.0

Introduces a global regularization method for 3D volume integral operators via a regularizing volume density interpolant derived from Green's identities, claiming high-order convergence including on curved domains.

Reference graph

Works this paper leans on

12 extracted references · cited by 1 Pith paper

[1]

SIAM Journal on Numerical Analysis26(5), 1212–1240 (1989)

Bernardi, C.: Optimal finite-element interpolation on curved domains. SIAM Journal on Numerical Analysis26(5), 1212–1240 (1989)

1989
[2]

North-Holland, Amsterdam (1978)

Ciarlet, P.G.: The Finite Element Method for Elliptic Problems. North-Holland, Amsterdam (1978)

1978
[3]

Computer Methods in Applied Mechanics and Engineering1, 217–249 (1972)

Ciarlet, P.G., Raviart, P.-A.: Interpolation theory over curved elements, with ap- plications to finite element methods. Computer Methods in Applied Mechanics and Engineering1, 217–249 (1972)

1972
[4]

Acta Numerica 22, 289–396 (2013)

Dziuk, G., Elliott, C.M.: Finite element methods for surface PDEs. Acta Numerica 22, 289–396 (2013)

2013
[5]

IMA Journal of Numerical Analysis33(2), 377–402 (2013)

Elliott, C.M., Ranner, T.: Finite element analysis for a coupled bulk–surface partial differential equation. IMA Journal of Numerical Analysis33(2), 377–402 (2013)

2013
[6]

SIAM, (2011)

Grisvard, P.: Elliptic Problems in Nonsmooth Domains. SIAM, (2011)

2011
[7]

Ishizaka, H., Kobayashi, K., Tsuchiya, T.: General theory of interpolation error estimates on anisotropic meshes. Jpn. J. Ind. Appl. Math.38(1), 163–191 (2021)

2021
[8]

Ishizaka, H., Kobayashi, K., Tsuchiya, T.: Anisotropic interpolation error estimates usinganewgeometricparameter.JapanJournalofIndustrialandAppliedMathematics 40, 475–512 (2023)

2023
[9]

Ishizaka, H., Kobayashi, K., Suzuki, R., Tsuchiya, T.: A new geometric condition equivalent to the maximum angle condition for tetrahedrons. Comput. Math. Appl. 99, 323–328 (2021)

2021
[10]

SIAM Journal on Numerical Analysis23(3), 562–580 (1986)

Lenoir, M.: Optimal isoparametric finite elements and error estimates for domains involving curved boundaries. SIAM Journal on Numerical Analysis23(3), 562–580 (1986)

1986
[11]

Scott, R.: Finite Element Techniques for Curved Boundaries. Ph.D. thesis, Massachu- setts Institute of Technology, Cambridge, MA (1973)

1973
[12]

Zlámal, M.: Curved elements in the finite element method. I. SIAM Journal on Numerical Analysis10(1), 229–240 (1973) Team FEM, Matsuyama, Japan Email address:h.ishizaka005@gmail.com

1973

[1] [1]

SIAM Journal on Numerical Analysis26(5), 1212–1240 (1989)

Bernardi, C.: Optimal finite-element interpolation on curved domains. SIAM Journal on Numerical Analysis26(5), 1212–1240 (1989)

1989

[2] [2]

North-Holland, Amsterdam (1978)

Ciarlet, P.G.: The Finite Element Method for Elliptic Problems. North-Holland, Amsterdam (1978)

1978

[3] [3]

Computer Methods in Applied Mechanics and Engineering1, 217–249 (1972)

Ciarlet, P.G., Raviart, P.-A.: Interpolation theory over curved elements, with ap- plications to finite element methods. Computer Methods in Applied Mechanics and Engineering1, 217–249 (1972)

1972

[4] [4]

Acta Numerica 22, 289–396 (2013)

Dziuk, G., Elliott, C.M.: Finite element methods for surface PDEs. Acta Numerica 22, 289–396 (2013)

2013

[5] [5]

IMA Journal of Numerical Analysis33(2), 377–402 (2013)

Elliott, C.M., Ranner, T.: Finite element analysis for a coupled bulk–surface partial differential equation. IMA Journal of Numerical Analysis33(2), 377–402 (2013)

2013

[6] [6]

SIAM, (2011)

Grisvard, P.: Elliptic Problems in Nonsmooth Domains. SIAM, (2011)

2011

[7] [7]

Ishizaka, H., Kobayashi, K., Tsuchiya, T.: General theory of interpolation error estimates on anisotropic meshes. Jpn. J. Ind. Appl. Math.38(1), 163–191 (2021)

2021

[8] [8]

Ishizaka, H., Kobayashi, K., Tsuchiya, T.: Anisotropic interpolation error estimates usinganewgeometricparameter.JapanJournalofIndustrialandAppliedMathematics 40, 475–512 (2023)

2023

[9] [9]

Ishizaka, H., Kobayashi, K., Suzuki, R., Tsuchiya, T.: A new geometric condition equivalent to the maximum angle condition for tetrahedrons. Comput. Math. Appl. 99, 323–328 (2021)

2021

[10] [10]

SIAM Journal on Numerical Analysis23(3), 562–580 (1986)

Lenoir, M.: Optimal isoparametric finite elements and error estimates for domains involving curved boundaries. SIAM Journal on Numerical Analysis23(3), 562–580 (1986)

1986

[11] [11]

Scott, R.: Finite Element Techniques for Curved Boundaries. Ph.D. thesis, Massachu- setts Institute of Technology, Cambridge, MA (1973)

1973

[12] [12]

Zlámal, M.: Curved elements in the finite element method. I. SIAM Journal on Numerical Analysis10(1), 229–240 (1973) Team FEM, Matsuyama, Japan Email address:h.ishizaka005@gmail.com

1973