pith. sign in

arxiv: 2606.01814 · v1 · pith:EIZPRSSRnew · submitted 2026-06-01 · 🧮 math.NA · cs.NA

The Immersed Discontinuous Galerkin Method for Elliptic Interface Problems

Lin Yang , Qilong Zhai This is my paper

Pith reviewed 2026-06-28 13:49 UTC · model grok-4.3

classification 🧮 math.NA cs.NA
keywords immersed finite elementdiscontinuous Galerkinelliptic interface problemsoptimal convergencetrace inequalitiescondition number robustness
0
0 comments X

The pith

Explicit linear immersed finite element functions satisfy interface jump conditions exactly without local solves and support an immersed discontinuous Galerkin scheme with optimal convergence rates.

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

The paper constructs linear immersed finite element functions for interface elements that match the exact jump conditions on the true interface. These functions are written in closed form and require no auxiliary problems or local linear systems to build. Rigorous analysis proves they retain optimal approximation power and satisfy trace inequalities whose constants stay bounded no matter how the interface slices the element. An immersed discontinuous Galerkin method assembled from them then delivers first-order H1 and second-order L2 convergence, while the stiffness-matrix condition numbers remain insensitive to the interface location inside each element.

Core claim

The constructed IFE functions achieve optimal approximation properties and satisfy the essential trace inequalities with constants independent of the interface cut; the resulting IDG scheme attains optimal convergence rates in the H1 and L2 norms, and the condition numbers of the stiffness matrices are robust with respect to the interface location.

What carries the argument

The explicit linear IFE functions that precisely satisfy the interface jump conditions on the actual interface in closed form.

If this is right

  • The IFE functions attain optimal approximation order in the energy norm with constants independent of the cut.
  • The immersed DG scheme converges at the optimal rates in both the H1 and L2 norms.
  • Stiffness-matrix condition numbers remain bounded independently of how the interface intersects the mesh.

Where Pith is reading between the lines

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

  • The same explicit-construction idea could be tested on quadratic or higher IFE spaces to check whether the independence from cut location persists.
  • Because no local solves are needed at each time step, the method may extend directly to moving-interface problems without repeated remeshing.

Load-bearing premise

The explicit construction of the linear IFE functions is assumed to be possible in closed form while exactly satisfying the interface jump conditions for arbitrary interface cuts through the element, without requiring additional local solves or auxiliary problems.

What would settle it

A sequence of successively refined meshes in which the observed H1 convergence rate drops below first order, or the stiffness-matrix condition number grows unboundedly, for at least one fixed family of interface cuts.

Figures

Figures reproduced from arXiv: 2606.01814 by Lin Yang, Qilong Zhai.

Figure 1
Figure 1. Figure 1: The transformation between the interface element [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The interface element. Lemma 3.1. [3] For the interface element T as shown in [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Two intersection patterns between the interface and an interface element. Left: Type 1 element, Right: [PITH_FULL_IMAGE:figures/full_fig_p016_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Example 7.1: The meshes of domain Ω with n = 4, 8, 16 [PITH_FULL_IMAGE:figures/full_fig_p023_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Example 7.1: The Condition numbers of the stiffness matrices. (Left: [PITH_FULL_IMAGE:figures/full_fig_p024_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Example 7.3: The meshes of domain Ω with n = 4, 8, 16 [PITH_FULL_IMAGE:figures/full_fig_p026_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Example 7.3: The Condition numbers of the stiffness matrices with [PITH_FULL_IMAGE:figures/full_fig_p027_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Example 7.4: The local partition with δ = 1/(40 ∗ 2 ℓ ). (Left: ℓ = 0, Middle: ℓ = 4, Right: ℓ = 8). 27 [PITH_FULL_IMAGE:figures/full_fig_p027_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Example 7.4: The local partition with δ = 1/20 − 1/(40 ∗ 2 ℓ ). (Left: ℓ = 1, Middle: ℓ = 3, Right: ℓ = 5). -1 0 1 2 3 4 5 6 7 8 9 10 ` 10 2 10 4 10 6 10 8 10 10 10 12 C o n ditio n N u m b er -1 = 1; -2 = 1 -1 = 1; -2 = 1000 -1 = 1000; -2 = 1 -1 0 1 2 3 4 5 6 7 8 9 10 ` 10 2 10 4 10 6 10 8 10 10 10 12 C o n ditio n N u m b er -1 = 1; -2 = 1 -1 = 1; -2 = 1000 -1 = 1000; -2 = 1 [PITH_FULL_IMAGE:figures/ful… view at source ↗
Figure 10
Figure 10. Figure 10: Example 7.4: The Condition numbers with n = 40. (Left: δ = 1/(40 ∗ 2 ℓ ), Right: δ = 1/20 − 1/(40 ∗ 2 ℓ )). In this example, we further investigate whether the condition numbers of the stiffness matrices depend on the position of the interface within an element, namely, whether they are affected by small-cut elements. As the parameter ℓ increases, the portion of subdomain Ω2 inside the interface element g… view at source ↗
read the original abstract

This paper is devoted to construction and convergence analysis of the linear explicit immersed finite element (IFE) function. For the interface elements, the proposed IFE functions precisely satisfy the interface conditions on the actual interface. The IFE functions are constructed in an explicit form and can be obtained directly without solving any auxiliary problems or local linear systems. Although the constructed IFE functions are non-polynomial, we establish rigorous theoretical analysis showing that they achieve optimal approximation properties and satisfy the essential trace inequalities. And the constants in the analysis are independent of how the interface cuts through the elements. Based on these IFE functions, an immersed discontinuous Galerkin numerical scheme is developed. Several numerical experiments are implemented to confirm that both the IFE functions and the numerical method achieve optimal convergence rates in the $H^1$ and $L^2$ norms. Furthermore, the numerical results indicate that the condition numbers of the stiffness matrices are robust with respect to the interface location.

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 / 1 minor

Summary. The paper constructs linear explicit immersed finite element (IFE) functions for elliptic interface problems. These functions exactly satisfy the jump conditions [u]=0 and [β ∂_n u]=0 on the true interface segment inside cut elements and are given in closed-form explicit (non-polynomial) expressions that require no auxiliary problems or local linear systems. The manuscript proves that the IFE functions attain optimal approximation properties and satisfy essential trace inequalities with constants independent of the interface cut position. An immersed discontinuous Galerkin (IDG) scheme is formulated from these functions; numerical experiments are reported to confirm optimal H¹ and L² convergence rates together with stiffness-matrix condition numbers that remain robust with respect to interface location.

Significance. If the explicit closed-form construction and the cut-independent analysis hold, the work supplies a practical advance for interface problems by eliminating per-element local solves while retaining optimal rates and robustness. This combination of explicit formulas, rigorous cut-independent estimates, and supporting numerics would strengthen the toolkit for elliptic problems with complex or moving interfaces.

major comments (2)
  1. [Construction of IFE functions] Abstract and construction section: the central claim that the linear IFE functions are obtained in explicit closed form while exactly enforcing both jump conditions for arbitrary cuts (including near-vertex configurations) is load-bearing for all subsequent approximation theory, trace inequalities, and conditioning results. The manuscript must exhibit the explicit expressions and demonstrate that they remain defined and stable without hidden divisions by quantities that vanish for certain cut geometries.
  2. [Approximation properties and trace inequalities] Analysis section on approximation properties: the proof that the approximation constants and trace inequalities are independent of the cut position must be checked for completeness; any step that implicitly depends on the distance from the interface to element vertices or edges would undermine the claimed cut-independence.
minor comments (1)
  1. [Numerical experiments] Numerical experiments section: the description of the test interfaces and the range of cut configurations examined should be expanded to make the robustness claim fully reproducible.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and for highlighting the potential practical value of the explicit IFE construction. We respond to each major comment below.

read point-by-point responses

  1. Referee: [Construction of IFE functions] Abstract and construction section: the central claim that the linear IFE functions are obtained in explicit closed form while exactly enforcing both jump conditions for arbitrary cuts (including near-vertex configurations) is load-bearing for all subsequent approximation theory, trace inequalities, and conditioning results. The manuscript must exhibit the explicit expressions and demonstrate that they remain defined and stable without hidden divisions by quantities that vanish for certain cut geometries.

    Authors: Section 3 derives and displays the explicit closed-form expressions for the linear IFE functions by directly enforcing [u]=0 and [β ∂_n u]=0 on the true interface segment inside each cut element. The resulting formulas are non-polynomial but closed-form, expressed solely in terms of the interface intersection coordinates and the coefficient β; no auxiliary problems or local linear systems are solved. For near-vertex cuts the denominators are geometric quantities (edge-segment lengths) that remain positive whenever the interface intersects two distinct edges, which is required for a valid interface element. We will add a short remark in Section 3 explicitly confirming that the expressions are defined and bounded for all admissible cuts, thereby making the stability statement more prominent. revision: partial

  2. Referee: [Approximation properties and trace inequalities] Analysis section on approximation properties: the proof that the approximation constants and trace inequalities are independent of the cut position must be checked for completeness; any step that implicitly depends on the distance from the interface to element vertices or edges would undermine the claimed cut-independence.

    Authors: Sections 4 and 5 prove optimal H¹ and L² approximation properties together with the required trace inequalities. The arguments rely on the exact satisfaction of the jump conditions by the explicit IFE functions, followed by standard polynomial approximation on the two sub-elements and inverse-trace estimates whose constants are controlled uniformly by the element diameter and the jump in β. All intermediate bounds are written so that they depend only on these quantities and never on the distance from the interface to a vertex or edge; the cut ratio appears only in terms that cancel or remain bounded. The proofs are therefore complete and cut-independent as stated. revision: no

Circularity Check

0 steps flagged

No significant circularity; explicit construction and independent analysis

full rationale

The paper presents an explicit closed-form construction of linear IFE functions that satisfy the interface jump conditions exactly on the true interface segment, followed by separate rigorous proofs of optimal approximation properties and cut-independent trace inequalities. No load-bearing steps reduce by definition or self-citation to fitted inputs; the central claims rest on direct construction plus mathematical analysis rather than renaming or circular re-use of results. This is the standard non-circular pattern for a numerical analysis paper deriving new basis functions and proving their properties.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The work relies on standard background assumptions from elliptic interface theory and finite element analysis; no free parameters, invented entities, or ad-hoc axioms are introduced in the abstract.

axioms (2)
  • domain assumption The interface is sufficiently regular (e.g., Lipschitz or C^1) so that the jump conditions are well-defined and the problem is well-posed.
    Standard prerequisite for convergence analysis of interface problems.
  • standard math The background mesh is shape-regular and quasi-uniform.
    Common assumption required for optimal approximation and trace inequalities in finite element theory.

pith-pipeline@v0.9.1-grok · 5692 in / 1356 out tokens · 30949 ms · 2026-06-28T13:49:07.902276+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

44 extracted references · 33 canonical work pages

  1. [1]

    Mu, Lin and Wang, Junping and Ye, Xiu , TITLE =. J. Comput. Appl. Math. , FJOURNAL =. 2015 , PAGES =. doi:10.1016/j.cam.2015.02.001 , URL =

    work page doi:10.1016/j.cam.2015.02.001 2015

  2. [2]

    Wang, Junping and Ye, Xiu , TITLE =. Adv. Comput. Math. , FJOURNAL =. 2016 , NUMBER =. doi:10.1007/s10444-015-9415-2 , URL =

    work page doi:10.1007/s10444-015-9415-2 2016

  3. [3]

    The finite element method for elliptic equations with discontinuous coefficients , JOURNAL =

    Babu. The finite element method for elliptic equations with discontinuous coefficients , JOURNAL =. 1970 , PAGES =. doi:10.1007/bf02248021 , URL =

    work page doi:10.1007/bf02248021 1970

  4. [4]

    and King, J

    Bramble, James H. and King, J. Thomas , TITLE =. Adv. Comput. Math. , FJOURNAL =. 1996 , NUMBER =. doi:10.1007/BF02127700 , URL =

    work page doi:10.1007/bf02127700 1996

  5. [5]

    Chen, Zhiming and Zou, Jun , TITLE =. Numer. Math. , FJOURNAL =. 1998 , NUMBER =. doi:10.1007/s002110050336 , URL =

    work page doi:10.1007/s002110050336 1998

  6. [6]

    and Li, Zhi Lin , TITLE =

    LeVeque, Randall J. and Li, Zhi Lin , TITLE =. SIAM J. Numer. Anal. , FJOURNAL =. 1994 , NUMBER =. doi:10.1137/0731054 , URL =

    work page doi:10.1137/0731054 1994

  7. [7]

    2006 , publisher=

    The immersed interface method: numerical solutions of PDEs involving interfaces and irregular domains , author=. 2006 , publisher=

    2006

  8. [8]

    and Massing, Andr\'

    Burman, Erik and Claus, Susanne and Hansbo, Peter and Larson, Mats G. and Massing, Andr\'. Cut. Internat. J. Numer. Methods Engrg. , FJOURNAL =. 2015 , NUMBER =. doi:10.1002/nme.4823 , URL =

    work page doi:10.1002/nme.4823 2015

  9. [9]

    Hansbo, Anita and Hansbo, Peter , TITLE =. Comput. Methods Appl. Mech. Engrg. , FJOURNAL =. 2002 , NUMBER =. doi:10.1016/S0045-7825(02)00524-8 , URL =

    work page doi:10.1016/s0045-7825(02)00524-8 2002

  10. [10]

    Mathematics of Computation , volume=

    A new multiscale finite element method for high-contrast elliptic interface problems , author=. Mathematics of Computation , volume=

  11. [11]

    SIAM Rev

    Hetmaniuk, Ulrich , TITLE =. SIAM Rev. , FJOURNAL =. 2011 , NUMBER =

    2011

  12. [12]

    Melenk, J. M. and Babu. The partition of unity finite element method: basic theory and applications , JOURNAL =. 1996 , NUMBER =. doi:10.1016/S0045-7825(96)01087-0 , URL =

    work page doi:10.1016/s0045-7825(96)01087-0 1996

  13. [13]

    International journal for numerical methods in engineering , volume=

    Partition of unity enrichment for bimaterial interface cracks , author=. International journal for numerical methods in engineering , volume=. 2004 , publisher=

    2004

  14. [14]

    An extended finite element method for modeling crack growth with frictional contact , JOURNAL =

    Dolbow, John and Mo\". An extended finite element method for modeling crack growth with frictional contact , JOURNAL =. 2001 , NUMBER =. doi:10.1016/S0045-7825(01)00260-2 , URL =

    work page doi:10.1016/s0045-7825(01)00260-2 2001

  15. [15]

    and Chopp, D

    Sukumar, N. and Chopp, D. L. and Mo\". Modeling holes and inclusions by level sets in the extended finite-element method , JOURNAL =. 2001 , NUMBER =. doi:10.1016/S0045-7825(01)00215-8 , URL =

    work page doi:10.1016/s0045-7825(01)00215-8 2001

  16. [16]

    Stable generalized finite element method (

    Babu. Stable generalized finite element method (. Comput. Methods Appl. Mech. Engrg. , FJOURNAL =. 2012 , PAGES =. doi:10.1016/j.cma.2011.09.012 , URL =

    work page doi:10.1016/j.cma.2011.09.012 2012

  17. [17]

    Strongly stable generalized finite element method: application to interface problems , JOURNAL =

    Babu. Strongly stable generalized finite element method: application to interface problems , JOURNAL =. 2017 , PAGES =. doi:10.1016/j.cma.2017.08.008 , URL =

    work page doi:10.1016/j.cma.2017.08.008 2017

  18. [18]

    Stable generalized finite element method and associated iterative schemes; application to interface problems , JOURNAL =

    Kergrene, Kenan and Babu. Stable generalized finite element method and associated iterative schemes; application to interface problems , JOURNAL =. 2016 , PAGES =. doi:10.1016/j.cma.2016.02.030 , URL =

    work page doi:10.1016/j.cma.2016.02.030 2016

  19. [19]

    Strongly stable generalized finite element method (

    Zhang, Qinghui and Banerjee, Uday and Babu. Strongly stable generalized finite element method (. Comput. Methods Appl. Mech. Engrg. , FJOURNAL =. 2019 , PAGES =. doi:10.1016/j.cma.2018.10.018 , URL =

    work page doi:10.1016/j.cma.2018.10.018 2019

  20. [20]

    Li, Zhilin , TITLE =. Appl. Numer. Math. , FJOURNAL =. 1998 , NUMBER =. doi:10.1016/S0168-9274(98)00015-4 , URL =

    work page doi:10.1016/s0168-9274(98)00015-4 1998

  21. [21]

    Li, Zhilin and Lin, Tao and Wu, Xiaohui , TITLE =. Numer. Math. , FJOURNAL =. 2003 , NUMBER =. doi:10.1007/s00211-003-0473-x , URL =

    work page doi:10.1007/s00211-003-0473-x 2003

  22. [22]

    Guo, Ruchi and Lin, Tao , TITLE =. IMA J. Numer. Anal. , FJOURNAL =. 2019 , NUMBER =. doi:10.1093/imanum/drx074 , URL =

    work page doi:10.1093/imanum/drx074 2019

  23. [23]

    SIAM Journal on Numerical Analysis

    An analysis of a broken P1-nonconforming finite element method for interface problems , author=. SIAM Journal on Numerical Analysis. , volume =

  24. [24]

    He, Xiaoming and Lin, Tao and Lin, Yanping , TITLE =. Numer. Methods Partial Differential Equations , FJOURNAL =. 2012 , NUMBER =. doi:10.1002/num.20620 , URL =

    work page doi:10.1002/num.20620 2012

  25. [25]

    Adjerid, Slimane and Ben-Romdhane, Mohamed and Lin, Tao , TITLE =. Int. J. Numer. Anal. Model. , FJOURNAL =. 2014 , NUMBER =

    2014

  26. [26]

    Guo, Ruchi and Lin, Tao , TITLE =. SIAM J. Numer. Anal. , FJOURNAL =. 2019 , NUMBER =. doi:10.1137/18M121318X , URL =

    work page doi:10.1137/18m121318x 2019

  27. [27]

    Adjerid, Slimane and Guo, Ruchi and Lin, Tao , TITLE =. Int. J. Numer. Anal. Model. , FJOURNAL =. 2017 , NUMBER =

    2017

  28. [28]

    Adjerid, Slimane and Lin, Tao and Meghaichi, Haroun , TITLE =. Comput. Methods Appl. Mech. Engrg. , FJOURNAL =. 2024 , PAGES =. doi:10.1016/j.cma.2023.116703 , URL =

    work page doi:10.1016/j.cma.2023.116703 2024

  29. [29]

    , TITLE =

    Ciarlet, Philippe G. , TITLE =. 1978 , PAGES =

    1978

  30. [30]

    Adjerid, Slimane and Lin, Tao and Meghaichi, Haroun , TITLE =. Comput. Methods Appl. Mech. Engrg. , FJOURNAL =. 2025 , NUMBER =. doi:10.1016/j.cma.2025.117829 , URL =

    work page doi:10.1016/j.cma.2025.117829 2025

  31. [31]

    , TITLE =

    Arnold, Douglas N. , TITLE =. SIAM J. Numer. Anal. , FJOURNAL =. 1982 , NUMBER =. doi:10.1137/0719052 , URL =

    work page doi:10.1137/0719052 1982

  32. [32]

    2008 , publisher=

    Discontinuous Galerkin methods for solving elliptic and parabolic equations: theory and implementation , author=. 2008 , publisher=

    2008

  33. [33]

    IEEE Trans

    Modeling Electrostatic Levitation of Dust Particles on Lunar Surface , AUTHOR=. IEEE Trans. Plasma Sci. , FJOURNAL=

  34. [34]

    1983 , PAGES =

    The mathematics of reservoir simulation , SERIES =. 1983 , PAGES =. doi:10.1137/1.9781611971071 , URL =

    work page doi:10.1137/1.9781611971071 1983

  35. [35]

    Numerical simulation of breakup of a viscous drop in simple shear flow through a volume-of-fluid method , AUTHOR=. Phys. Fluids , issue=

  36. [36]

    Ji, Haifeng and Li, Zhilin , TITLE =. J. Sci. Comput. , FJOURNAL =. 2026 , NUMBER =. doi:10.1007/s10915-025-03155-x , URL =

    work page doi:10.1007/s10915-025-03155-x 2026

  37. [37]

    Ji, Haifeng , TITLE =. J. Sci. Comput. , FJOURNAL =. 2022 , NUMBER =. doi:10.1007/s10915-022-01839-2 , URL =

    work page doi:10.1007/s10915-022-01839-2 2022

  38. [38]

    ESAIM Math

    Ji, Haifeng and Wang, Feng and Chen, Jinru and Li, Zhilin , TITLE =. ESAIM Math. Model. Numer. Anal. , FJOURNAL =. 2023 , NUMBER =. doi:10.1051/m2an/2023047 , URL =

    work page doi:10.1051/m2an/2023047 2023

  39. [39]

    Adjerid, Slimane and Chaabane, Nabil and Lin, Tao , TITLE =. Comput. Methods Appl. Mech. Engrg. , FJOURNAL =. 2015 , PAGES =. doi:10.1016/j.cma.2015.04.006 , URL =

    work page doi:10.1016/j.cma.2015.04.006 2015

  40. [40]

    Camp, Brian and Lin, Tao and Lin, Yanping and Sun, Weiwei , TITLE =. Adv. Comput. Math. , FJOURNAL =. 2006 , NUMBER =. doi:10.1007/s10444-004-4139-8 , URL =

    work page doi:10.1007/s10444-004-4139-8 2006

  41. [41]

    and Lin, Y

    Lin, T. and Lin, Y. and Sun, W.-W. and Wang, Z. , TITLE =. J. Comput. Appl. Math. , FJOURNAL =. 2011 , NUMBER =. doi:10.1016/j.cam.2011.01.041 , URL =

    work page doi:10.1016/j.cam.2011.01.041 2011

  42. [42]

    Lin, Tao and Lin, Yanping and Zhang, Xu , TITLE =. SIAM J. Numer. Anal. , FJOURNAL =. 2015 , NUMBER =. doi:10.1137/130912700 , URL =

    work page doi:10.1137/130912700 2015

  43. [43]

    Lin, Tao and Sheen, Dongwoo and Zhang, Xu , TITLE =. J. Comput. Phys. , FJOURNAL =. 2013 , PAGES =. doi:10.1016/j.jcp.2013.03.053 , URL =

    work page doi:10.1016/j.jcp.2013.03.053 2013

  44. [44]

    Lin, Tao and Yang, Qing and Zhang, Xu , TITLE =. J. Sci. Comput. , FJOURNAL =. 2015 , NUMBER =. doi:10.1007/s10915-015-9989-3 , URL =

    work page doi:10.1007/s10915-015-9989-3 2015