Pith. sign in

REVIEW 2 major objections 6 minor 1 cited by

FIAT: enabling classical and modern macroelements

T0 review · 2 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper claims that FIAT, the finite element tabulator, now supports a broad, general-purpose family of macroelements — cells subdivided into smaller pieces — and that this support is integrated into the Firedrake software stack.

desk verdict Solid, well-scoped software contribution that delivers a general macroelement facility for FIAT/Firedrake with explicit transformation theory and reproducible convergence tests; send it to a serious referee. read the letter →

arxiv 2501.14599 v2 pith:BJYZKXNM submitted 2025-01-24 math.NA cs.NA

Pablo D. Brubeck , Robert C. Kirby This is my paper · ORCID
classification math.NAcs.NA MSC 65N3065D07
keywords macroelementsFIATFiredrakefiniteelementtabulationC1continuitydivergence-freeStokesHCTsimplicialcomplexes
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper sets out to make macroelements a first-class, general feature of the FIAT finite element library and the Firedrake code stack. Macroelements work by splitting each computational cell into smaller pieces, which lets finite element spaces enforce properties such as $C^1$ continuity or a pointwise divergence-free velocity at lower polynomial degree than ordinary elements. The authors claim that their new machinery covers a wide range of classical and modern elements — HCT and higher-order HCT, Powell-Sabin, Scott-Vogelius, Guzmán–Neilan, Alfeld–Sorokina, and Johnson–Mercier — and that numerical tests reproduce the theoretically predicted convergence rates. If the claim is right, researchers can request such spaces as off-the-shelf building blocks instead of hand-coding each element.

What carries the argument

The carrying mechanism is the reference-element macroelement pipeline: a macroelement is defined by a space of piecewise polynomials over a splitting of the reference simplex and a list of dual functionals, with the nodal basis obtained by solving a generalized Vandermonde system. For elements whose degrees of freedom involve derivatives, the key identity is the factored nodal-transformation matrix $V = E V_c D$, built from a 'completion' of the node set; it maps physical nodes to reference nodes even when push-forwards of derivative functionals do not align with reference functionals.

What would settle it

Run a biharmonic test with a Powell-Sabin 6 element defined through the reference pipeline on a triangulation where adjacent triangles form a nonconvex union; if the affine-split assumption is doing the work, the $C^1$ continuity across internal edges will fail and the $H^2$ error will stop converging at first order on that mesh.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that a macroelement can be treated as an ordinary Ciarlet triple over a reference simplicial complex: define the piecewise-polynomial space on the split reference cell, list functionals as degrees of freedom, and let FIAT's Vandermonde construction produce the nodal basis, while a companion transformation theory handles degrees of freedom such as normal derivatives that are not preserved by simple pullback. The implementation introduces a SimplicialComplex class, a SplitSimplicialComplex encoding Alfeld, iso, Powell-Sabin-12 and Wang splittings, a MacroQuadrature that tiles quadrature over subcells, and integration through FInAT and TSFC into Firedrake. The paper reports that the resulting elements attain their expected convergence orders for Stokes flow, the Navier-Stokes cylinder benchmark, and the biharmonic plate problem.

Load-bearing premise

The whole pipeline assumes every physical cell subdivision is the affine image of a fixed reference subdivision, so geometrically-dependent splits such as incenter-based Powell-Sabin 6 or Worsey-Farin tetrahedral splits fall outside the framework as presented.

Editorial extensions

If this is right

  • Users of Firedrake can request macroelement spaces such as HCT, Powell-Sabin, or Lagrange with an 'alfeld' or 'iso' variant directly from the FunctionSpace constructor.
  • The divergence-free Stokes pairs (Scott-Vogelius, Guzmán-Neilan, Alfeld-Sorokina) and the Johnson-Mercier stress element attain their predicted convergence rates, with the divergence-free pairs keeping the discrete incompressibility residual near machine precision.
  • Element assembly cost for macroelements scales as $N_{dof}^2 N_q^{ref} |\Delta(K)|$; some elements such as PS12 and HCT4 are as expensive to assemble as classical high-order elements, though solve time dominates total run time.
  • Per the authors, this is the first general-purpose implementation of such a macroelement facility, going beyond one-off support in other finite element packages.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural extension the paper leaves implicit is to lift the affine-split restriction by allowing the reference-to-physical map itself to carry the geometry of the split, which would bring incenter-based Powell-Sabin 6 and Worsey-Farin tetrahedral elements into the same framework.
  • The flattened quadrature strategy in TSFC could be reworked as an inner loop over subcells; this would likely cut assembly cost for iso-type $C^0$ elements, whose basis functions are locally supported on few subcells.
  • The nodal-completion transformation theory is general enough that the same factored pattern should apply to other derivative-based macroelements, for instance high-order Argyris-type spaces built on Alfeld splits.
  • Because the implementation records parent-to-child relationships in the simplicial complex, it may provide a natural place to define the inter-grid transfers needed for multigrid solvers on macroelement spaces.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 6 minor

Summary. This paper describes an extension of the FIAT finite element tabulator and the Firedrake code stack to support simplicial macroelements: elements defined on a regular subdivision of each cell. The authors introduce infrastructure in FIAT for representing split simplices and composite quadrature, integrate this with FInAT and TSFC so that macroelement spaces can be used through Firedrake's FunctionSpace interface, and apply the transformation theory of Kirby (2018) to construct the HCT macroelement and its reduced and higher-order variants. They implement a suite of Stokes-relevant macroelements (Scott--Vogelius, Guzmán--Neilan, Alfeld--Sorokina, Johnson--Mercier) and C1 elements (Powell--Sabin, HCT, reduced HCT), and document convergence studies for Stokes flow in two and three dimensions, the Navier--Stokes cylinder benchmark, and the biharmonic plate problem. The paper claims this is the first general-purpose macroelement facility of this breadth, and the code is archived on Zenodo.

Significance. If the implementation is correct, the paper fills a real gap in general-purpose finite element software: previously, macroelements such as HCT, Powell--Sabin, Scott--Vogelius, and Johnson--Mercier were supported only as one-off elements in individual packages, if at all. The main technical contribution, an explicit derivation of the HCT transformation matrices from the general theory in Kirby (2018), is self-contained and parameter-free. The numerical results are also convincing: they cover several independent benchmark problems and report the theoretically expected convergence rates, and the manuscript provides a Zenodo archive of the exact software and scripts, which makes the experiments independently checkable. The authors are also explicit about the main limitation, namely that the reference-element pipeline is restricted to affinely preserved splits based on barycenters; this is an honest scope statement rather than a hidden defect. The work should be of interest both to users of Firedrake and to researchers developing macroelement discretizations.

major comments (2)
  1. [Section 2.5 / Section 5.1] The description of the three-dimensional Alfeld--Sorokina (AS) results is ambiguous in a way that affects the validity of a central numerical claim. Section 2.5 states that the quadratic Alfeld--Sorokina macroelement is only inf-sup stable in two dimensions, and that inf-sup stability in three dimensions requires enriching it with the Guzmán--Neilan cubic divergence-free face bubbles. However, Section 5.1 lists the AS pair simply as "the Alfeld--Sorokina velocity element paired with C0 P1 pressure on the Alfeld split," and Figure 10 reports three-dimensional convergence results labeled "AS" without stating whether the enriched space was used. If the 3D experiments use the enriched element, the degrees of freedom and the relationship to the unenriched AS space should be stated; if they use the unenriched space, the paper needs to explain why the inf-sup condition holds in these examples. Without this clarification, the 3D AS convergence results cannot be interpreted.
  2. [Section 2.6] The stated dimension formula for the Johnson--Mercier space is incorrect as written. The text says the space has dimension \binom{d+1}{2} d(d+1), which gives 18 for d=2 and 72 for d=3, whereas the paper states (correctly, from the listed degrees of freedom) that the dimensions are 15 and 42. The correct count is d(d+1)(2d+1)/2, obtained from the facet moments and the integral averages described in that section. This should be corrected because the dimension count is part of the motivation for using the Johnson--Mercier element and because the erroneous formula would mislead anyone implementing the element from the description.
minor comments (6)
  1. [Section 5.1] The sentence describing the Alfeld--Sorokina pair says it is "paired with C0 P1 velocity on the Alfeld split"; this should presumably read "paired with C0 P1 pressure." The current wording makes the discretization nonsensical.
  2. [Section 4.1] In the paragraph containing equation (4.19), "and smooth function f = f∘F" should be "and smooth function \hat f = f∘F" (or similar), since the symbol f is used for the physical function elsewhere.
  3. [Section 4.1] The definition of B_i in equation (4.21) uses the symbol G without a subscript, while the surrounding text defines G_i for each edge; this is a notational slip that should be fixed to avoid confusion about which edge's normal/tangent frame is being used.
  4. [Figures 9 and 10] The captions of Figures 9 and 10 state that "the other formulations use macro-elements based on the Alfeld split," but the ISO element uses the uniform iso-split rather than the Alfeld split; the captions should be corrected or qualified.
  5. [Section 5.3] The sentence "The macroelements give lower orders of accuracy, but have fewer global degrees of freedom and lower polynomial degree than the Bell and Argyris elements" is not true for HCT4: Table 4 shows HCT4 has 995 global rows on the 8x8 mesh, while Bell has 486 and Argyris has 694. The statement should be restricted to the lower-order macroelements or rephrased to acknowledge the HCT4 case.
  6. [Section 4.1] The treatment of the reduced HCT transformation is described only as being "quite analogous" to constructing the Bell transformation from the Argyris transformation. Since the reduced HCT element is tested numerically in Section 5.3, a short outline of the completion and extraction procedure for this case would improve reproducibility, even if a full derivation is not necessary.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the macroelement transformation theory is re-derived in the paper and the numerical benchmarks are external.

full rationale

The paper's central claim is a software-engineering one: FIAT and Firedrake can now construct, transform, and evaluate a broad class of affinely-invariant simplicial macroelements. There is no fitted parameter anywhere in the derivation chain, and no 'prediction' is produced from a fit. The convergence studies in Section 5 compare against known theoretical rates and against published reference values for the cylinder flow problem (e.g., Nabh 1998), so the numerical results are independently checkable. The transformation theory of Section 4 is imported from Kirby 2018, a prior parameter-free general result whose stated assumptions do not include the macroelement targets; the paper then explicitly re-derives the specific HCT transformation matrices D, V_c, and E in Eqs. (4.13)-(4.24), so the application is not asserted by self-citation alone. The higher-order HCT construction is supported by its own derivations using Jacobi orthogonality and integration by parts, Eqs. (4.28)-(4.29). The acknowledged limitation to affinely-preserved barycentric splits is stated openly in Section 2 and delimits the claimed scope rather than being smuggled in as an unstated assumption. Prior self-citations, such as FIAT 2004 and Kirby 2018, are normal building blocks and are not load-bearing in the sense of forcing the paper's numerical outcomes. No circular step is identifiable under the specified criteria.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The central claim rests on standard finite element theory, the affine-invariance of the implemented splits, and the previously published transformation theory of Kirby 2018. No fitted parameters are introduced, and no new physical entities are postulated. The only design choices are documented (Jacobi polynomial bases, Xiao-Gimbutas quadrature rules).

assumptions (3)
  • standard math Ciarlet triple and unisolvence determine the nodal basis via the generalized Vandermonde system (Eq. 2.4)
    Underpins the FIAT approach of pairing a basis of the approximating space with a basis of the dual space; used throughout Section 3 for constructing macroelement bases.
  • domain assumption Each macroelement splitting is affinely invariant: the physical-cell split is the image of the reference split under the cell affine map
    Stated in Section 2: 'Our development concentrates on affinely-preserved splits based on barycenters.' This is the load-bearing limitation of the framework; non-affinely-preserved splits (incenter-based PS6, Worsey-Farin tetrahedral splits) are explicitly out of scope.
  • standard math The transformation theory of Kirby 2018 applies to macroelements with derivative degrees of freedom
    The pullback/push-forward and nodal completion machinery of Kirby 2018 is assumed valid and is applied to HCT and reduced HCT in Section 4. This is prior published work with stated assumptions, not a circular dependency.

how reviews work

0 comments
Cite this review

Pith. "Pith review of FIAT: enabling classical and modern macroelements." pith.science (2026). https://pith.science/paper/BJYZKXNM

@misc{pith2026250114599,
  author       = {Pith},
  title        = {Pith review of: FIAT: enabling classical and modern macroelements},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BJYZKXNM}},
  note         = {Machine review of arXiv:2501.14599}
}
abstract

Many classical and modern finite element spaces are derived by dividing each computational cell into finer pieces. Such \emph{macroelements} frequently enable the enforcement of mathematically desirable properties such as divergence-free conditions or $C^1$ continuity in a simpler or more efficient manner than elements without the subdivision. Although a few modern software projects provide one-off support for particular macroelements, a general approach facilitating broad-based support has, until now, been lacking. In this work, we describe a major addition to the FIAT project to support a wide range of different macroelements. These enhancements have been integrated into the Firedrake code stack. We provide numerical evaluation of the new macroelement facility.

Figures

Figures reproduced from arXiv: 2501.14599 by the authors.

Figure 1
Figure 1. Some typical splitting strategies for macroelements. [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Some Lagrange-type 𝐶 0 macroelements. Solid dots represent point evaluation degrees of freedom. 2.4 𝐶 1 macroelements For 𝐶 1 elements, we focus on splittings of a triangle 𝐾 rather than general simplex. For some splitting Δ of 𝐾, we define 𝑆 1 𝑘 (Δ) =  𝑠 ∈ 𝐶 1 (𝐾) : 𝑠|𝜏 ∈ P𝑘 (𝜏), 𝜏 ∈ Δ [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Quadratic 𝐶 1 macroelements on Powell-Sabin splits. Hollow circles represent derivative evaluation for each Cartesian direction, and arrows represent normal derivative moments along edges. (a) Reduced HCT (b) HCT3 (c) HCT4 [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (16 more)
Figure 4
Figure 4. Figure 4: HCT-type 𝐶 1 macroelements on the Alfeld split. Solid lines represent moments along edges. The 12-dimensional HCT space, 𝑆 1 3 (Δ𝐴 (𝐾)), can be parametrized in the same way as 𝑆 1 2 (Δ𝑃𝑆12 (𝐾)), and the 9-dimensional reduced HCT space ˜𝑆3 (Δ𝐴 (𝐾)) as 𝑆 1 2 (Δ𝑃𝑆6 (𝐾)). …
Figure 5
Figure 5. Figure 5: Lowest-order macroelements for the Stokes and elasticity complexes. The thin arrows in Guzmán– [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Affine mapping to a reference cell 𝐾ˆ from a typical cell 𝐾. Note that here 𝐹 maps from the physical cell 𝐾 to the reference cell 𝐾ˆ rather than the other way around. As shown in [Kirby 2018], we have that 𝑀 = 𝑉 ⊤. When the function space is preserved under pullback an…
Figure 7
Figure 7. Figure 7: Pushing forward the HCT derivative nodes in physical space does [PITH_FULL_IMAGE:figures/full_fig_p013_7.png]
Figure 8
Figure 8. Figure 8: Nodal sets 𝑁ˆ 𝑐 and 𝑁 𝑐 giving the compatible nodal completion of 𝑁 and 𝑁ˆ for an HCT element and reference element are formed by including tangential derivatives along with normal derivatives at each edge midpoint. The space is 12-dimensional, so the lengths of N and …
Figure 9
Figure 9. Figure 9: Convergence plots for various Stokes discretizations on an [PITH_FULL_IMAGE:figures/full_fig_p018_9.png]
Figure 10
Figure 10. Figure 10: Convergence plots for various Stokes discretizations on an [PITH_FULL_IMAGE:figures/full_fig_p019_10.png]
Figure 11
Figure 11. Figure 11: FLOP count for evaluating the kernels for computing the elementwise contributions to matrices [PITH_FULL_IMAGE:figures/full_fig_p019_11.png]
Figure 12
Figure 12. Figure 12: Sparsity patterns for several Stokes pairs on an [PITH_FULL_IMAGE:figures/full_fig_p020_12.png]
Figure 13
Figure 13. Figure 13: Timing results versus the number of degrees of freedom for the two-dimensional Stokes equations [PITH_FULL_IMAGE:figures/full_fig_p021_13.png]
Figure 14
Figure 14. Figure 14: Computational domain for flow past cylinder, with boundary conditions indicated on each part of [PITH_FULL_IMAGE:figures/full_fig_p021_14.png]
Figure 15
Figure 15. Figure 15: Error in drag, lift, pressure drop, and divergence for the 2d cylinder problem. Here, the Johnson– [PITH_FULL_IMAGE:figures/full_fig_p022_15.png]
Figure 16
Figure 16. Figure 16: Error in solving biharmonic equation on a perturbed [PITH_FULL_IMAGE:figures/full_fig_p023_16.png]
Figure 17
Figure 17. Figure 17: FLOP count for evaluating the element-level kernel for the biharmonic operator. This count assumes [PITH_FULL_IMAGE:figures/full_fig_p024_17.png]
Figure 18
Figure 18. Figure 18: Sparsity patterns of the discrete biharmonic operator on a regular [PITH_FULL_IMAGE:figures/full_fig_p025_18.png]
Figure 19
Figure 19. Figure 19: Timing results versus the number of degrees of freedom for the two-dimensional biharmonic equation [PITH_FULL_IMAGE:figures/full_fig_p025_19.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. Automated Runge-Kutta-Nystr\"om time stepping for finite element methods in Irksome

    math.NA 2025-08 conditional novelty 5.0 of 10

    Irksome now automates Runge-Kutta-Nyström time stepping for second-order PDEs in Firedrake, producing smaller stage-coupled systems and faster solves than first-order Runge-Kutta methods.

Reference graph

Works this paper leans on

47 extracted references · 35 canonical work pages · cited by 1 Pith paper

  1. [14]

    to appear, ACM Transactions on Mathematical Software (2024)

    FIAT: improving performance and accuracy for high-order finite elements. to appear, ACM Transactions on Mathematical Software (2024). Carsten Carstensen, Joscha Gedicke, and Eun-Jae Park

    work page 2024

  2. [15]

    xviii+397 pages

    Springer, New York. xviii+397 pages. Susanne C Brenner and Li-Yeng Sung. 2005.𝐶0 interior penalty methods for fourth order elliptic boundary value problems on polygonal domains. Journal of Scientific Computing 22, 1 (2005), 83–118. Pablo D. Brubeck, Robert C. Kirby, Fabian Laakmann, and Lawrence Mitchell

    work page 2005

  3. [16]

    arXiv preprint arXiv:2403.13189 (2024)

    The Johnson-Mercier elasticity element in any dimensions. arXiv preprint arXiv:2403.13189 (2024). Jay Gopalakrishnan, Philip L Lederer, and Joachim Schöberl

    arXiv 2024

  4. [18]

    Computer Methods in Applied Mechanics and Engineering 395 (2022), 114983

    Generalized 𝐶1 Clough–Tocher splines for CAGD and FEM. Computer Methods in Applied Mechanics and Engineering 395 (2022), 114983. https://doi.org/10.1016/j.cma.2022.114983 Johnny Guzmán and Michael Neilan

    work page arXiv 2022

  5. [21]

    New development in FreeFem++. J. Numer. Math. 20, 3-4 (2012), 251–265. https://freefem.org/ Miklós Homolya, Robert C. Kirby, and David A. Ham

    work page 2012

  6. [23]

    SIAM Journal on Scientific Computing 40, 3 (2018), C401–C428

    TSFC: a structure-preserving form compiler. SIAM Journal on Scientific Computing 40, 3 (2018), C401–C428. Volker John

    work page 2018

  7. [26]

    ACM Trans

    Algorithm 839: FIAT, a new paradigm for computing finite element basis functions. ACM Trans. Math. Software 30, 4 (2004), 502–516. https://doi.org/10.1145/1039813.1039820 Robert C. Kirby

    work page arXiv 2004

  8. [29]

    Fast simplicial finite element algorithms using Bernstein polynomials. Numer. Math. 117, 4 (2011), 631–652. Robert C. Kirby

    work page 2011

Show all 47 references
  1. [30]

    SIAM Journal on Scientific Computing 36, 2 (2014), A846–A868

    Low-complexity finite element algorithms for the de Rham complex on simplices. SIAM Journal on Scientific Computing 36, 2 (2014), A846–A868. Robert C. Kirby

    2014

  2. [31]

    SMAI Journal of Computational Mathematics 4 (2018), 197–224

    A general approach to transforming finite elements. SMAI Journal of Computational Mathematics 4 (2018), 197–224. https://doi.org/10.5802/smai-jcm.33 Robert C. Kirby and Anders Logg

    2018 doi

  3. [32]

    ACM Trans

    A compiler for variational forms. ACM Trans. Math. Software 32, 3 (2006), 417–444. Robert C. Kirby and Lawrence Mitchell

    2006

  4. [34]

    SIAM Journal on Scientific Computing 38, 5 (2016), S25–S47

    Automated generation and symbolic manipulation of tensor product finite elements. SIAM Journal on Scientific Computing 38, 5 (2016), S25–S47. https://doi.org/10.1137/15M1021167 arXiv:1411.2940 [math.NA] Lin Mu, Xiu Ye, and Shangyou Zhang

    2016 arXiv

  5. [35]

    Numerical Methods for Partial Differential Equations 35, 4 (2019), 1497–1508

    Development of a𝑃2 element with optimal𝐿2 convergence for biharmonic equation. Numerical Methods for Partial Differential Equations 35, 4 (2019), 1497–1508. Guido Nabh

    2019

  6. [36]

    SIAM Journal on Scientific Computing 45, 2 (2023), A675–A702

    Low-order preconditioning for the high-order finite element de Rham complex. SIAM Journal on Scientific Computing 45, 2 (2023), A675–A702. Michael J. D. Powell and Malcolm A. Sabin

    2023

  7. [38]

    ACM Trans

    Firedrake: automating the finite element method by composing abstractions. ACM Trans. Math. Software43, 3 (2016), 24:1–24:27. https://doi.org/10.1145/2998441 arXiv:1501.01809 [cs.MS] Yves Renard and Konstantinos Poulios

    2016 arXiv

  8. [39]

    ACM Trans

    GetFEM: Automated FE modeling of multiphysics problems based on a generic weak form language. ACM Trans. Math. Software 47, 1 (2020), 1–31. Marie E. Rognes, Robert C. Kirby, and Anders Logg

    2020

  9. [40]

    SIAM Journal on Scientific Computing 31, 6 (2010), 4130–4151

    Efficient assembly of 𝐻(div)and 𝐻(curl)conforming fi- nite elements. SIAM Journal on Scientific Computing 31, 6 (2010), 4130–4151. https://doi.org/10.1137/08073901X arXiv:1205.3085 [math.NA] M. Schäfer, S. Turek, F. Durst, E. Krause, and R. Rannacher

    2010 arXiv

  10. [42]

    Journal of Open Source Software 7, 73 (2022),

    Basix: a runtime finite element basis evaluation library. Journal of Open Source Software 7, 73 (2022),

    2022

  11. [44]

    Internat

    𝐶1 macroelements in adaptive finite element methods. Internat. J. Numer. Methods Engrg. 70, 9 (2007), 1076–1095. Andrew J Worsey and B Piper

    2007

  12. [46]

    Computers & Mathematics with Applications 59, 2 (2010), 663–676

    A numerical algorithm for the construction of efficient quadrature rules in two and higher dimensions. Computers & Mathematics with Applications 59, 2 (2010), 663–676. https://doi.org/10.1016/j. camwa.2009.10.027 zenodo/Zenodo-20250826.1

    2010 doi

  13. [1977]

    ACM Transactions on Mathematical Software (TOMS) 3, 4 (1977), 316–325

    Piecewise quadratic approximations on triangles. ACM Transactions on Mathematical Software (TOMS) 3, 4 (1977), 316–325. Florian Rathgeber, David A. Ham, Lawrence Mitchell, Michael Lange, Fabio Luporini, Andrew T. T. McRae, Gheorghe-Teodor Bercea, Graham R. Markall, and Paul H....

    1977

  14. [1978]

    Some equilibrium finite element methods for two-dimensional elasticity problems. Numer. Math. 30 (1978), 103–116. George Karniadakis and Spencer J Sherwin

    1978

  15. [1984]

    Japan Journal of Applied Mathematics 1, 2 (1984), 347–367

    PEERS: a new mixed finite element for plane elasticity. Japan Journal of Applied Mathematics 1, 2 (1984), 347–367. Douglas N. Arnold and Jinshui Qin

    1984

  16. [1985]

    Analysis of some finite elements for the Stokes problem. Math. Comp. 44 (1985), 71–79. https://doi.org/10.1090/S0025-5718-1985-0771031-7 Andreas Bock, Colin J. Cotter, and Robert C. Kirby

    1985 doi

  17. [1988]

    Computer Aided Geometric Design 5, 3 (1988), 177–186

    A trivariate Powell-Sabin interpolant. Computer Aided Geometric Design 5, 3 (1988), 177–186. Hong Xiao and Zydrunas Gimbutas

    1988

  18. [1992]

    Advances in computer methods for partial differential equations 7 (1992), 28–34

    Quadratic velocity/linear pressure Stokes elements. Advances in computer methods for partial differential equations 7 (1992), 28–34. Douglas N. Arnold and Ragnar Winther

    1992

  19. [1996]

    Vieweg+Teubner Verlag, Wiesbaden, 547–566

    Benchmark Computations of Laminar Flow Around a Cylinder. Vieweg+Teubner Verlag, Wiesbaden, 547–566. https://doi.org/10.1007/978-3-322-89849-4_39 L. Ridgway Scott and Micheal Vogelius. 1984.Conforming Finite Element Methods for Incompressible and Nearly Incompressible Continua...

    1984 doi

  20. [2002]

    Mixed finite elements for elasticity. Numer. Math. 92, 3 (Sept. 2002), 401–419. https://doi.org/10.1007/s002110100348 Douglas N. Arnold and Ragnar Winther

    2002 doi

  21. [2003]

    Mathematical Models and Methods in Applied Sciences 13, 03 (March 2003), 295–307

    Nonconforming mixed elements for elasticity. Mathematical Models and Methods in Applied Sciences 13, 03 (March 2003), 295–307. https://doi.org/10.1142/s0218202503002507 Francis R. A. Aznaran, Patrick E. Farrell, and Robert C. Kirby

    2003 doi

  22. [2004]

    International Journal for Numerical Methods in Fluids 44, 7 (2004), 777–788

    Reference values for drag and lift of a two-dimensional time-dependent flow around a cylinder. International Journal for Numerical Methods in Fluids 44, 7 (2004), 777–788. Claes Johnson and Bertrand Mercier

    2004

  23. [2006]

    ACM Trans

    Optimizing FIAT with level 3 BLAS. ACM Trans. Math. Software 32, 2 (2006), 223–235. https: //doi.org/10.1145/1141885.1141889 FIAT: enabling classical and modern macroelements 27 Robert C. Kirby

    2006

  24. [2007]

    II—a general-purpose object-oriented finite element library

    deal. II—a general-purpose object-oriented finite element library. ACM Transactions on Mathematical Software (TOMS) 33, 4 (2007), 24–es. Christine Bernardi and Genevivève Raugel

    2007

  25. [2008]

    Finite elements for symmetric tensors in three dimensions. Math. Comp. 77, 263 (July 2008), 1229–1251. https://doi.org/10.1090/S0025-5718-08-02071-1 Douglas N. Arnold, Gerard Awanou, and Ragnar Winther

    2008 doi

  26. [2010]

    ACM Trans

    Singularity-free evaluation of collapsed-coordinate orthogonal polynomials. ACM Trans. Math. Software 37, 1 (2010), 1–16. https://doi.org/10.1145/1644001.1644006 Robert C. Kirby

    2010

  27. [2011]

    SIAM Journal on Scientific Computing 33, 6 (2011), 3087–3109

    Bernstein–Bézier finite elements of arbitrary order and optimal assembly procedures. SIAM Journal on Scientific Computing 33, 6 (2011), 3087–3109. Peter Alfeld and Tatyana Sorokina

    2011

  28. [2012]

    SIAM Journal on Scientific Computing 34, 4 (2012), A2267–A2287

    Numerical experiments for the Arnold–Winther mixed finite elements for the Stokes problem. SIAM Journal on Scientific Computing 34, 4 (2012), A2267–A2287. Philippe G Ciarlet

    2012

  29. [2014]

    Mathematical Models and Methods in Applied Sciences 24, 04 (2014), 783–796

    Nonconforming tetrahedral mixed finite elements for elasticity. Mathematical Models and Methods in Applied Sciences 24, 04 (2014), 783–796. https://doi.org/10.1142/ s021820251350067x 26 Pablo D. Brubeck and Robert C. Kirby Douglas N Arnold, Franco Brezzi, and Jim Douglas

    2014

  30. [2016]

    BIT Numerical Mathematics 56 (2016), 15–32

    Linear differential operators on bivariate spline spaces and spline vector fields. BIT Numerical Mathematics 56 (2016), 15–32. https://doi.org/10.1007/s10543-015-0557-x Douglas N. Arnold, Gerard Awanou, and Ragnar Winther

    2016 doi

  31. [2017]

    arXiv:1711.02473 [cs.MS] Miklós Homolya, Lawrence Mitchell, Fabio Luporini, and David A Ham

    Exposing and exploiting structure: optimal code generation for high-order finite element methods. arXiv:1711.02473 [cs.MS] Miklós Homolya, Lawrence Mitchell, Fabio Luporini, and David A Ham

    arXiv

  32. [2018]

    Inf-sup stable finite elements on barycentric refinements producing divergence– free approximations in arbitrary dimensions. SIAM J. Numer. Anal. 56, 5 (2018), 2826–2844. David A. Ham, Paul H. J. Kelly, Lawrence Mitchell, Colin J. Cotter, Robert C. Kirby, Koki Sagiyama, Nacime...

    2018

  33. [2019]

    ACM Trans

    Code generation for generally mapped finite elements. ACM Trans. Math. Software 45, 4 (2019), 41:1–41:23. https://doi.org/10.1145/3361745 arXiv:1808.05513 [cs.MS] Ming-Jun Lai and Larry L Schumaker

    2019 arXiv

  34. [2020]

    A mass conserving mixed stress formulation for Stokes flow with weakly imposed stress symmetry. SIAM J. Numer. Anal. 58, 1 (2020), 706–732. Jan Grošelj and Marjeta Knez

    2020

  35. [2022]

    The SMAI Journal of computational mathematics 8 (2022), 399–437

    Transformations for Piola-mapped elements. The SMAI Journal of computational mathematics 8 (2022), 399–437. https://doi.org/10.5802/smai-jcm.91 arXiv:2110.13224 [math.NA] Wolfgang Bangerth, Ralf Hartmann, and Guido Kanschat

    2022 arXiv

  36. [2023]

    Imperial College London and University of Oxford and Baylor University and University of Washington

    Firedrake User Manual (first ed.). Imperial College London and University of Oxford and Baylor University and University of Washington. https://doi.org/10.25561/104839 F. Hecht

    doi

  37. [2024]

    Computers & Mathematics with Applications 159 (2024), 155–172

    Planar curve registration using Bayesian inversion. Computers & Mathematics with Applications 159 (2024), 155–172. https://doi.org/10.1016/j.camwa.2024.02.005 arXiv:2307.04909 [cs.CV] Susanne C. Brenner and L. Ridgway Scott

    2024 arXiv

  38. [2025]

    https://doi.org/10.5281/zenodo.16954245

    Software used in ‘Code for FIAT: enabling classical and modern macroelements (rev 1)’. https://doi.org/10.5281/zenodo.16954245

    doi

  39. [3982]

    Stogner and Graham F

    https://doi.org/10.21105/joss.03982 Roy H. Stogner and Graham F. Carey

    doi

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.