pith. sign in

arxiv: 2506.11897 · v4 · submitted 2025-06-13 · 🧮 math.NA · cs.NA

A multiphase cubic MARS method for fourth- and higher-order interface tracking of two or more materials with arbitrary topology and geometry

Yan Tan , Yixiao Qian , Zhiqi Li , Qinghai Zhang This is my paper

Pith reviewed 2026-05-19 09:27 UTC · model grok-4.3

classification 🧮 math.NA cs.NA
keywords interface trackingmultiphasecubic splineshigh-order accuracyarbitrary topologyjunctionsnumerical methods
0
0 comments X

The pith

A multiphase cubic MARS method tracks interfaces of arbitrary numbers of materials at fourth- through eighth-order accuracy in space and time.

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

The paper proposes a multiphase cubic MARS method that represents the topology and geometry of interfaces between any number of materials using graphs, cycles, and cubic splines. It maintains a user-specified regularity condition on marker spacing while adapting marker density to local curvature. All junction types are treated uniformly without special cases. The approach reaches fourth-, sixth-, and eighth-order accuracy both temporally and spatially. This level of precision supports detailed simulations where interface evolution controls the outcome.

Core claim

The multiphase cubic MARS method represents the topology and geometry of the interface via graphs, cycles, and cubic splines, maintains an (r,h)-regularity of the interface, distributes the markers adaptively along the interface so that arcs with high curvature are resolved by densely populated markers, and achieves fourth-, sixth-, and eighth-order accuracy both in time and in space for interface tracking of an arbitrary number of materials with arbitrarily complex topology and geometry, while handling all possible types of junctions with ease.

What carries the argument

Cubic spline representation of interfaces defined by graphs and cycles, combined with maintenance of (r,h)-regularity and adaptive marker distribution.

Load-bearing premise

The interface regularity and adaptive marker placement along cubic splines can be preserved for arbitrarily complex topologies without introducing errors that reduce the claimed high-order accuracy, particularly near junctions.

What would settle it

A benchmark test with multiple junctions and regions of high curvature in which the observed convergence rate falls below fourth order would show the accuracy claim does not hold.

Figures

Figures reproduced from arXiv: 2506.11897 by Qinghai Zhang, Yan Tan, Yixiao Qian, Zhiqi Li.

Figure 1
Figure 1. Figure 1: Boundary representations of multiple phases. Subplot (a) shows three adjacent phases to be [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The boundary representation of six pairwise disjoint Yin sets whose regularized union covers [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The pipeline of constructing Γ as a spline approximation of Γ, cf. Notation 4.26. e concatenating marker sequences in EX in (4.3) according to circuits or trails in CS ∪ TS and then interpolating each concatenated breakpoint sequence: a periodic spline for a circuit and a not-a-knot spline for a trail (that is not a circuit). Cut SCT at junctions and kinks and we obtain the spline edge set SE in Defini￾tio… view at source ↗
Figure 4
Figure 4. Figure 4: Algorithm 2 enforces rb > 1 for the image of a not-a-knot spline s n under the discrete flow map φ k tn . In subplot (a), the three knots at the starting end p0 of φ k tn (s n) cause rb < 1, where p0 = Sφ(l0), p1 = Sφ(l1), and Sφ : [l0, l1] → R2 is given by Sφ := φ k tn ◦ s n. As shown in Step (ARMS-4b) of Definition 5.4, this undesirable situation is fixed by first ensuring that the chordal length ∥p0 − p… view at source ↗
Figure 5
Figure 5. Figure 5: The ARMS strategy. In subplot (a), the interface markers on [PITH_FULL_IMAGE:figures/full_fig_p030_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Fitting a cubic spline on a marker sequence is equivalent to adding markers from the [PITH_FULL_IMAGE:figures/full_fig_p036_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: The removal of a marker pi from the breakpoint sequence (pi)N i=0 of a cubic spline is equivalent to perturbing pi to ˆpi on the new spline as the output of χj , which, fitted through the new sequence (p0, p1, . . . , pi−1, pi+1, . . . , pN ), is C 3 at ˆpi. where l is the cumulative chordal length determined from the breakpoint sequence (Xk ) where (Xt ) :=  ϕ˚t 0 (X0 i ) N i=0 . Proof. At any time t ∈ … view at source ↗
Figure 7
Figure 7. Figure 7: For any sufficiently small k > 0, the marker sequence of ∂φk tj−1Mj−1 ψ sat￾isfies the (r, hL)-regularity for some r ∈ (0, rtiny]. Then Lemma 4.20, Lemma 6.3, and the fact of the error over [li−1, li+1] around the perturbed marker pi being O(hL) · O(h 4 L ) = O(h 5 L ) imply [PITH_FULL_IMAGE:figures/full_fig_p039_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Solutions of the cubic MARS method for the vortex shear test with [PITH_FULL_IMAGE:figures/full_fig_p045_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Solutions of the cubic MARS method with h c L = 0.2h for the deformation test of T = 2 and T = 4 on the Eulerian grid of h = 1 32 . Subplots (a)–(f) are snapshots of the solution at key time instants. In subplots (g)–(j), each phase is represented by a curve of the same color except that the white phase (the unbounded complement of the circle) is represented by the black curve. Due to symmetry, the red and… view at source ↗
Figure 10
Figure 10. Figure 10: Solutions of the cubic MARS method with h c L = 0.2h for the vortex shear test (T = 16) and the deformation test (T = 4) on the Eulerian grid of h = 1 32 . Subplot (a) shows the initial Yin sets, which are the same as those in [PITH_FULL_IMAGE:figures/full_fig_p053_10.png] view at source ↗
read the original abstract

For interface tracking of an arbitrary number of materials in two dimensions, we propose a multiphase cubic MARS method that (a) represents the topology and geometry of the interface via graphs, cycles, and cubic splines, (b) applies to any number of materials with arbitrarily complex topology and geometry, (c) maintains an $(r,h)$-regularity of the interface so that the distance between any pair of adjacent markers is within a user-specified range, (d) distributes the markers adaptively along the interface so that arcs with high curvature are resolved by densely populated markers, and (e) achieves fourth-, sixth-, and eighth-order accuracy both in time and in space.} In particular, all possible types of junctions, which pose challenges to VOF methods and level-set methods, are handled with ease. Results of a variety of benchmark tests confirm the analysis and demonstrate the superior accuracy, efficiency, and versatility of the proposed method.

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 proposes a multiphase cubic MARS method for interface tracking of an arbitrary number of materials in two dimensions. The interface topology and geometry are represented via graphs, cycles, and cubic splines. The method maintains (r,h)-regularity so that adjacent marker distances lie in a user-specified range, distributes markers adaptively according to curvature, and is claimed to achieve fourth-, sixth-, and eighth-order accuracy in both time and space. All junction types are asserted to be handled without difficulty, in contrast to VOF and level-set approaches. A variety of benchmark tests are presented to support the analysis and to illustrate accuracy, efficiency, and versatility.

Significance. If the claimed high-order convergence rates are substantiated by error analysis and numerical evidence that the order is preserved even at junctions for arbitrarily complex topologies, the work would constitute a useful contribution to high-order interface tracking methods, offering a graph-based alternative that avoids the topological difficulties of other Eulerian schemes.

major comments (2)
  1. [Abstract and §3] Abstract and §3 (method description, features (c) and (d)): The central claim of sixth- and eighth-order spatial accuracy rests on the assertion that (r,h)-regularity enforcement and curvature-adaptive cubic-spline redistribution introduce no order reduction at junctions. Cubic splines furnish at most fourth-order interpolation; any C^1 projection or redistribution step therefore risks injecting O(h^4) local errors. The manuscript must supply either a truncation-error analysis isolating junction neighborhoods or numerical convergence tables that separately report L^∞ and L^2 errors inside small balls around triple and higher junctions. Without such evidence the global high-order claim is not yet load-bearing.
  2. [§4 or §5] §4 or §5 (error analysis or numerical results): The abstract states that “results of a variety of benchmark tests confirm the analysis,” yet no table or figure isolates the observed convergence rate in the vicinity of junctions. If the reported orders are measured only on smooth interface segments away from junctions, they do not yet corroborate the claim that “all possible types of junctions … are handled with ease” while retaining sixth- or eighth-order accuracy.
minor comments (2)
  1. [Title and Abstract] The title and abstract should explicitly state that the method is formulated in two dimensions; the phrase “two or more materials” alone does not convey the spatial setting.
  2. [§3] Notation for the user-specified range in the (r,h)-regularity condition should be introduced once and used consistently; the current description leaves the precise definition of the range ambiguous.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading of our manuscript and the constructive comments on the high-order accuracy claims. We address each major comment below and will strengthen the presentation with additional analysis and numerical evidence in the revised version.

read point-by-point responses

  1. Referee: [Abstract and §3] Abstract and §3 (method description, features (c) and (d)): The central claim of sixth- and eighth-order spatial accuracy rests on the assertion that (r,h)-regularity enforcement and curvature-adaptive cubic-spline redistribution introduce no order reduction at junctions. Cubic splines furnish at most fourth-order interpolation; any C^1 projection or redistribution step therefore risks injecting O(h^4) local errors. The manuscript must supply either a truncation-error analysis isolating junction neighborhoods or numerical convergence tables that separately report L^∞ and L^2 errors inside small balls around triple and higher junctions. Without such evidence the global high-order claim is not yet load-bearing.

    Authors: We appreciate the referee pointing out the need for explicit verification that the redistribution steps preserve the claimed orders at junctions. The method combines cubic-spline geometry representation with a high-order time integrator and a carefully formulated (r,h)-regularity constraint whose local truncation error is designed to be consistent with the target order. Nevertheless, we agree that a dedicated truncation-error analysis focused on junction neighborhoods would make the argument rigorous. In the revised manuscript we will add such an analysis together with numerical tables that report L^∞ and L^2 errors inside small balls centered at triple and higher junctions. revision: yes

  2. Referee: [§4 or §5] §4 or §5 (error analysis or numerical results): The abstract states that “results of a variety of benchmark tests confirm the analysis,” yet no table or figure isolates the observed convergence rate in the vicinity of junctions. If the reported orders are measured only on smooth interface segments away from junctions, they do not yet corroborate the claim that “all possible types of junctions … are handled with ease” while retaining sixth- or eighth-order accuracy.

    Authors: We agree that global error measures alone do not fully isolate the behavior at junctions. The benchmark suite already contains test cases with multiple triple and higher junctions of arbitrary topology, and the reported convergence rates are computed over the entire interface. To directly address the referee’s request, we will include additional tables and figures in the revised manuscript that separately extract and tabulate observed orders using only the markers lying inside small neighborhoods of the junctions. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation is self-contained algorithmic construction

full rationale

The abstract and described features present the multiphase cubic MARS method as an independent proposal: interface topology via graphs/cycles/cubic splines, (r,h)-regularity maintenance, curvature-adaptive marker redistribution, and explicit claims of fourth/sixth/eighth-order accuracy in time and space. No equations, fitted parameters, or self-citations are exhibited that reduce the accuracy orders or junction-handling claims to definitions, tautologies, or prior self-referential results by construction. Benchmark tests are invoked as external verification rather than internal fits. The central claims therefore remain non-circular and rest on the method's stated algorithmic properties rather than reducing to their inputs.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The abstract provides limited explicit axioms or parameters; the method implicitly relies on standard numerical assumptions for spline interpolation and marker advection.

free parameters (1)
  • user-specified range for adjacent marker distance
    The (r,h)-regularity condition uses a user-chosen interval to control marker spacing; this is an input parameter rather than a fitted constant.
axioms (1)
  • domain assumption Interfaces of arbitrary topology can be represented and evolved using graphs, cycles, and cubic splines while preserving high-order accuracy.
    Invoked throughout the listed features (a)-(e) as the foundation for handling complex junctions and adaptive resolution.

pith-pipeline@v0.9.0 · 5714 in / 1324 out tokens · 40819 ms · 2026-05-19T09:27:50.943901+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

33 extracted references · 33 canonical work pages

  1. [1]

    V. I. Arnold, Ordinary Differential Equations (MIT Press, 1973), iSBN: 0-262-51018-9

    work page 1973

  2. [2]

    D. J. Benson, Computational methods in Lagrangian and Eulerian hydrocodes, Com- put. Methods Appl. Mech. Eng. 99 (1992) 235–394

    work page 1992

  3. [3]

    D. J. Benson, Eulerian finite element methods for the micromechanics of heterogeneous materials: Dynamic prioritization of material interfaces,Comput. Methods Appl. Mech. Eng. 151 (1998) 343–360

    work page 1998

  4. [4]

    D. J. Benson, Volume of fluid interface reconstruction methods for multi-material problems, Appl. Mech. Rev. 55 (2002) 151–165

    work page 2002

  5. [5]

    D. Bonn, J. Eggers, J. Indekeu, J. Meunier and E. Rolley, Wetting and spreading, Reviews of Modern Physics 81 (2009) 739–805

    work page 2009

  6. [6]

    Caboussat, M

    A. Caboussat, M. M. Francois, R. Glowinski, D. B. Kothe and J. M. Sicilian, A numerical method for interface reconstruction of triple points within a volume tracking algorithm, Mathematical and Computer Modelling 48 (2008) 1957–1971

    work page 2008

  7. [7]

    B. Y. Choi and M. Bussmann, A piecewise linear approach to volume tracking a triple point, Int. J. Numer. Meth. Fluids 53 (2007) 1005–1018

    work page 2007

  8. [8]

    de Niem, E

    D. de Niem, E. K¨ uhrt and U. Motschmann, A volume-of-fluid method for simulation of compressible axisymmetric multi-material flow, Comput. Phys. Commun. 176 (2007) 170–190

    work page 2007

  9. [9]

    Diestel, Graph Theory (Springer, 2017), fifth edition

    R. Diestel, Graph Theory (Springer, 2017), fifth edition

    work page 2017

  10. [10]

    J. R. Dormand and P. J. Prince, High order embedded Runge-Kutta formulae, J. Comput. Appl. Math. 7 (1981) 67–75

    work page 1981

  11. [11]

    Dyadechko and M

    V. Dyadechko and M. Shashkov, Reconstruction of multi-material interfaces from moment data, J. Comput. Phys. 227 (2008) 5361–5384

    work page 2008

  12. [12]

    Hergibo, T

    P. Hergibo, T. N. Phillips and Z. Xie, A moment-of-fluid method for resolving fila- mentary structures using a symmetric multi-material approach,J. Comput. Phys. 491 (2023) 112401

    work page 2023

  13. [13]

    C. W. Hirt and B. D. Nichols, Volume of fluid (VOF) method for the dynamics of free boundaries, J. Comput. Phys. 39 (1981) 201–225

    work page 1981

  14. [14]

    D. Hu, K. Liang, L. Ying, S. Li and Q. Zhang, ARMS: Adding and removing markers on splines for high-order general interface tracking under the MARS framework, J. Comput. Phys. 521 (2025) 113574

    work page 2025

  15. [15]

    Jemison, M

    M. Jemison, M. Sussman and M. Shashkov, Filament capturing with the multimaterial moment-of-fluid method, J. Comput. Phys. 285 (2015) 149–172

    work page 2015

  16. [16]

    Kucharik, R

    M. Kucharik, R. V. Garimella, S. P. Schofield and M. J. Shashkov, A comparative August 26, 2025 9:3 WSPC/INSTRUCTION FILE MARS-n2D 56 Y. Tan & Y. Qian & Z. Li & Q. Zhang study of interface reconstruction methods for multi-material ALE simulations,J. Com- put. Phys. 229 (2010) 2432–2452

    work page 2025

  17. [17]

    Mosso and S

    S. Mosso and S. Clancy, A geometrically derived priority system for Young’s interface reconstruction, Technical report, Los Alamos National Laboratory, Los Alamos, NM, 1995

    work page 1995

  18. [18]

    Osher and J

    S. Osher and J. A. Sethian, Fronts propagating with curvature-dependent speed: Al- gorithms based on Hamilton-Jacobi formulations, J. Comput. Phys. 79 (1988) 12–49

    work page 1988

  19. [19]

    E. G. Puckett, A volume-of-fluid interface tracking algorithm with applications to computing shock wave refraction, in Proceedings of the Fourth International Syposium on Computational Fluid Dynamics , ed. H. Dwyer (1991), pp. 933–938

    work page 1991

  20. [20]

    R. I. Saye and J. A. Sethian, The Voronoi implicit interface method for computing multiphase physics, PNAS 108 (2011) 19498–19503

    work page 2011

  21. [21]

    S. P. Schofield, R. V. Garimella, M. M. Francois and R. Loub` ere, A second-order accu- rate material-order-independent interface reconstruction technique for multi-material flow simulations, J. Comput. Phys. 228 (2009) 731–745

    work page 2009

  22. [22]

    Sijoy and S

    C. Sijoy and S. Chaturvedi, Volume-of-fluid algorithm with different modified dynamic material ordering methods and their comparisons, J. Comput. Phys. 229 (2010) 3848– 3863

    work page 2010

  23. [23]

    Y. Sui, H. Ding and P. D. Spelt, Numerical simulations of flows with moving contact lines, Annu. Rev. Fluid Mech. 46 (2014) 97–119

    work page 2014

  24. [24]

    Tryggvason, B

    G. Tryggvason, B. Bunner, D. Juric, W. Tauber, S. Nas, J. Han, N. Al-Rawahi. and Y.- J. Jan, A front–tracking method for the computations of multiphase flow, J. Comput. Phys. 169 (2001) 708–759

    work page 2001

  25. [25]

    L. W. Tu, An Introduction To Manifolds (Springer Nature, New York, 2011), 2nd edition

    work page 2011

  26. [26]

    J. H. Verner, Explicit Runge–Kutta methods with estimates of the local truncation error, SIAM J. Numer. Anal. 15 (1978) 772–790

    work page 1978

  27. [27]

    D. L. Youngs, Time dependent multi-material flow with large fluid distortion, in Nu- merical Methods for Fluid Dynamics, eds. K. W. Morton and M. J. Baines (1982), pp. 273–285

    work page 1982

  28. [28]

    Zhang, GePUP: Generic projection and unconstrained PPE for fourth-order solu- tions of the incompressible Navier-Stokes equations with no-slip boundary conditions, J

    Q. Zhang, GePUP: Generic projection and unconstrained PPE for fourth-order solu- tions of the incompressible Navier-Stokes equations with no-slip boundary conditions, J. Sci. Comput. 67 (2016) 1134–1180

    work page 2016

  29. [29]

    Zhang, Fourth-and higher-order interface tracking via mapping and adjusting reg- ular semianalytic sets represented by cubic splines, SIAM J

    Q. Zhang, Fourth-and higher-order interface tracking via mapping and adjusting reg- ular semianalytic sets represented by cubic splines, SIAM J. Sci. Comput. 40 (2018) A3755–A3788

    work page 2018

  30. [30]

    Zhang and A

    Q. Zhang and A. Fogelson, MARS: An analytic framework of interface tracking via mapping and adjusting regular semialgebraic sets, SIAM J. Numer. Anal. 54 (2016) 530–560

    work page 2016

  31. [31]

    Zhang and Z

    Q. Zhang and Z. Li, Boolean algebra of two-dimensional continua with arbitrarily complex topology, Math. Comput. 89 (2020) 2333–2364

    work page 2020

  32. [32]

    Zhang, Y

    Q. Zhang, Y. Zhu and Z. Li, An AI-aided algorithm for multivariate polynomial recon- struction on Cartesian grids and the PLG finite difference method, J. Sci. Comput. 101 (2024) 66

    work page 2024

  33. [33]

    Q. Zhao, W. Ren and Z. Zhang, A thermodynamically consistent model and its con- servative numerical approximation for moving contact lines with soluble surfactants, Comput. Methods Appl. Mech. Engrg. 385 (2021) 114033

    work page 2021