Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

Monotonicity and convergence of two-relaxation-times lattice Boltzmann schemes for a non-linear conservation law

T0 review · 3 major / 5 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 Under two explicit inequalities on the relaxation parameters, two-relaxation-times lattice Boltzmann schemes for scalar conservation laws are monotone and converge to the weak entropy solution as the grid is refined.

desk verdict A solid, genuinely new TRT extension of the BGK convergence analysis, but the central proof is not fully self-contained in the preprint. read the letter →

arxiv 2501.07934 v2 pith:CK6R45H7 submitted 2025-01-14 math.NA cs.NA

Denise Aregba-Driollet (IMB , Bordeaux INP) , Thomas Bellotti (EM2C) This is my paper · ORCID
classification math.NAcs.NA MSC 65M1265M0865M0635L60
keywords two-relaxation-timeslatticeBoltzmannschememonotonicityscalarconservationlawentropysolutionL1-contractivityconvergencenumericaldiffusionkinetic
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 prove that two-relaxation-times (TRT) lattice Boltzmann schemes—the middle ground between simple BGK and fully general multi-relaxation-times schemes—are convergent for scalar non-linear conservation laws. It gives explicit conditions on the two relaxation parameters $\omega_s$ and $\omega_a$ under which the collide-and-stream operator is monotone in the distribution functions, and then follows the monotone-scheme route: invariant compact set, $L^1$-contractivity, total variation bounds, closeness to equilibrium, and a weak-form test-function argument that identifies the limit as the unique weak entropy solution. The point of TRT is that one parameter can be raised toward 2, which lowers numerical diffusion, while the other keeps monotonicity, so the scheme can be sharper than BGK and still provably convergent. Numerical experiments on D1Q3 and D2Q5 lattices confirm the predicted convergence and the reduced diffusion.

What carries the argument

The load-bearing object is the relaxation operator written in symmetric/anti-symmetric variables: each pair of opposed velocities $c_{2\ell}=-c_{2\ell+1}$ is split into $s_{2\ell}=(f_{2\ell}+f_{2\ell+1})/2$ and $a_{2\ell}=(f_{2\ell}-f_{2\ell+1})/2$, relaxed at rates $\omega_s$ and $\omega_a$. With equilibria of the form $f_i^{eq}=L_i u + \sum_k N_{i,k}\varphi_k(u)$ and $N_{2\ell,k}=-N_{2\ell+1,k}\ge0$, the Jacobian entries of the relaxation are affine in the flux derivatives, and requiring them non-negative on the invariant box gives exactly conditions (3.5)–(3.6). Monotonicity then yields the full convergence toolbox: invariant set, $\ell^1$-contractivity of the relaxation (Proposition 4.4), $L^1$-contractivity and total-variation decay, an $O(\Delta x)$ estimate for the distance to equilibrium (Proposition 4.8), and finally the weak-form limit argument that produces the entropy inequality.

What would settle it

Run the D1Q3 test with flux $\varphi(u)=u^2/2$, the magic combination $\omega_s=50/73$, $\omega_a=96/73$ (inside the monotonicity set $M$), and initial data $u_0(x)=1_{[0,1/2]}(|x|)$ together with a small shift of the same data; the paper predicts the discrete $L^1$ distance between the two runs never increases, so a single time step where it grows refutes the $L^1$-contractivity claim that underpins the convergence theorem.

Watch

Extended reading notes

Core claim

The central claim is that a TRT lattice Boltzmann scheme for $\partial_t u + \sum_{k=1}^d \partial_{x_k}\varphi_k(u)=0$ converges to the unique weak entropy solution, provided the equilibria have the structural form $f_i^{eq}(u)=L_i u + \sum_k N_{i,k}\varphi_k(u)$ with the linear part symmetric along each link and the flux part anti-symmetric, and provided the relaxation parameters satisfy $\omega_s L_1 \ge \max(0,\omega_s-1)$ and the per-link bound $\omega_a \max_{u\in[-u_\infty,u_\infty]}|\sum_k N_{2\ell,k}\varphi_k'(u)| \le \omega_s L_{2\ell} + \tfrac12 \min(2-\omega_s-\omega_a,0,\omega_a-\omega_s)$. Under these conditions the relaxation operator is monotone non-decreasing on an invariant set of distribution functions, and the scheme inherits $L^\infty$ bounds, $L^1$ contractivity, total-variation decay, geometric convergence to equilibrium, and ultimately strong $L^1$ convergence to the unique entropy solution.

Load-bearing premise

The proof rests on the structural assumption that every equilibrium distribution is an affine function of the conserved quantity plus a weighted copy of the flux, with the flux term entirely in the anti-symmetric part and the linear term entirely in the symmetric part.

Editorial extensions

If this is right

  • TRT schemes can run with $\omega_a>1$, and under the magic combination $\omega_s+\omega_a=2$ even close to 2, while remaining provably monotone and convergent; the BGK scheme, with one parameter, must stay much closer to 1.
  • Any scheme satisfying (3.5)–(3.6) keeps all distribution functions inside their equilibrium box and the conserved moment inside $[-u_\infty,u_\infty]$ for all times.
  • Two numerical solutions whose initial data differ by $\|u_0-v_0\|_{L^1}$ never grow apart in $L^1$, and the total variation of each solution never exceeds that of the initial datum.
  • The distribution functions approach their equilibrium at a geometric rate in time and stay within $O(\Delta x)$ of equilibrium, so the numerical modes become slaves of the physical mode as the grid refines.
  • The limit $u$ of the conserved moment satisfies both the weak form of the conservation law and the kinetic entropy inequalities for every constant $\kappa$, which identifies it as the unique entropy solution.

Reading between the lines

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

  • Editorial extension: because the paper notes the monotonicity conditions are sufficient but not necessary, a natural next experiment is to scan pairs $(\omega_s,\omega_a)$ outside $M$ and measure the entropy error; finding parameter regions outside $M$ with vanishing error would indicate where the sufficient conditions can be widened.
  • Editorial extension: the structural assumption (3.1) is the main restriction; the paper remarks that the flux only needs to be locally Lipschitz, so the same proof should extend to fluxes with kinks such as $\varphi(u)=|u|$ with minor technical changes.
  • Editorial extension: since the entropy argument is scalar, extending TRT convergence to systems of conservation laws would require a different mechanism, but the same link-wise monotonicity conditions could be checked against known finite-volume entropy stability criteria.
  • Editorial extension: a quantitative test of the sharpness of (4.5) would be to track $\|f^n_\Delta - f^{eq}(u^n_\Delta)\|_{L^1}$ for many random initial data and check whether the $O(\Delta x \operatorname{TV}(u_0))$ bound and the geometric decay predicted by (4.6) hold with the computed constant.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper studies two-relaxation-times (TRT) lattice Boltzmann schemes for multidimensional scalar conservation laws with smooth fluxes and BV initial data. Under an affine-in-the-conserved-moment ansatz for the equilibria (Eq. (3.1) with the symmetry assumptions (3.3)), the authors derive monotonicity conditions on the relaxation parameters (Proposition 3.3, conditions (3.5)--(3.6)). They then use monotonicity to prove invariance of a compact set, L1-contractivity, total-variation bounds, and closeness-to-equilibrium estimates (Propositions 4.2--4.8). These estimates are assembled into a Crandall--Majda--Lax--Wendroff type argument (Theorems 4.10 and 4.11) claiming convergence, along a subsequence, to the unique weak entropy solution of the conservation law. Numerical experiments for D1Q3 and D2Q5 schemes illustrate the monotonicity region M and report convergence rates.

Significance. If the deferred proofs are correct, the paper gives the first convergence analysis covering the full TRT parameter range for scalar conservation laws, improving on the BGK-specific results of reference [1]. The explicit monotonicity conditions (3.5)--(3.6) are concrete and testable, and they depend on the equilibrium coefficients L_i and N_{i,k} in a way that practitioners can check. A notable strength is that the main estimates visible in the main text (Propositions 3.3, 4.4, 4.5, 4.7, 4.8) are algebraically explicit and internally consistent. The paper is also honest about the scope of its assumptions: the affine equilibrium ansatz (3.1)--(3.3) is stated up front and is not hidden. The numerical experiments, including mesh-refinement tables and invariant-compact-set checks, support the qualitative claim that TRT schemes can reduce numerical diffusion while preserving monotonicity.

major comments (3)
  1. [§4.1.5, Theorem 4.10, Eq. (4.8)] The convergence to a weak solution is not proven in the main text: after deriving the balance D - I = F, the proof stops at 'We are left to handle F ... but can be treated by analogous arguments—see supplementary material.' The term F contains precisely the lattice Boltzmann collision/streaming contribution that must be converted into the flux integral of the weak form. The omitted step requires showing that finite differences of shifted test functions against R_i(f_Delta) produce the flux term and that the remainder vanishes, using Proposition 4.8 and the Lipschitz continuity of R on K. Without this derivation, Theorem 4.10 is conditional on material not present in the manuscript.
  2. [§4.2, Theorem 4.11] The entropy inequality is justified only formally: the sentence 'Formally (rigorous justifications ... can be obtained as for Theorem 4.10)' covers the limit passage in the discrete Kruzhkov entropy balance. Since Theorem 4.10 itself has an unproved flux limit, the entropy solution claim inherits the same gap. The main text thus does not establish convergence to the entropy solution, which is the paper's headline result.
  3. [§3.2, Propositions 3.9, 3.10, 3.11] Three results that are load-bearing for the subsequent estimates are stated with proofs in supplementary material: Proposition 3.9 (monotonicity of the equilibria and the characterization of M_BGK) is used in the proof of Proposition 4.5 to justify the L1-contractivity bound and the TV bound; Proposition 3.10 (ωs ≠ 2 and ωa ≠ 2 for monotone schemes) is used in Proposition 4.8 to ensure that max(|1−ωs|,|1−ωa|) < 1; Proposition 3.11 is used for the geometric description of M. The main text therefore relies on unproved results for its core stability and convergence estimates.
minor comments (5)
  1. [Title] The title as rendered contains stray spaces ('RELAXA TION-TIMES LA TTICE BOL TZMANN'); these are clearly formatting artifacts and should be fixed.
  2. [§2.2.2, Eq. (2.3)] The relaxation operator R_i is defined on R^q, but the collision step is written for each velocity index i; it might help to state explicitly that R_i depends on the full vector (f_1,...,f_q) and that the Jacobian entries are evaluated at generic points in K, not only at equilibrium.
  3. [§3.2, conditions (3.5)--(3.6)] The term min(2−ωs−ωa, 0, ωa−ωs) in (3.6) is not immediately transparent; a short explanation of how it arises from taking the minimum of the three right-hand sides in (3.8)--(3.10) would help readability.
  4. [§5.1.3, Table 1] For the case ωa = 2 with hat initial data, the reported empirical order is about 1.3, but the row with ∆x = 2.44e-04 is followed by a jump to 1.76e+13 in Table 2 for the BGK case; this is presumably an instability artifact and should be briefly explained in the caption or text.
  5. [References] References [24] and [23] are cited for monotonicity-preserving linear multistep methods; the connection to the lattice Boltzmann context in Remark 3.5 would be clearer if the specific properties used were indicated.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the TRT monotonicity and convergence results follow from explicit structural and monotonicity assumptions, and the deferred flux-limit passages are completeness gaps rather than circular reductions.

full rationale

Proposition 3.3 derives conditions (3.5)-(3.6) directly from the non-negativity of the Jacobian of the relaxation operator under the explicit equilibrium ansatz (3.1)/(3.3), and Proposition 4.4 derives ℓ1-contractivity from the same conditions. The convergence proof in Theorem 4.10 then combines the invariant compact set, TV estimates, L1-contractivity, and closeness-to-equilibrium (Proposition 4.8) with a Lax-Wendroff summation-by-parts argument; the only deferred part is the treatment of the term F in equation (4.8), which the text states 'can be treated by analogous arguments—see supplementary material'. Similarly, Theorem 4.11 says 'Formally (rigorous justifications ... can be obtained as for Theorem 4.10)'. These are omitted proofs, not circular inputs: the flux limit is not assumed as the target flux integral, it is asserted to follow from a separate argument, and the entropy limit is likewise presented as obtainable by the same route. The paper's self-citations ([1], [3], [5]-[7]) supply the BGK baseline, modified-equation theory, and finite-difference reformulation, but they do not define the TRT convergence claim into existence. The equilibrium split (3.1)/(3.3) is an explicit, testable assumption rather than a conclusion smuggled in by citation. No fitted parameter is renamed as a prediction, and no known result is merely relabeled. The main limitation is incompleteness of the main text, not circularity.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the structural equilibrium split (3.1)-(3.3), which restricts the class of TRT schemes, plus standard well-posedness and initialization assumptions. No free parameters are fitted to data; the relaxation parameters omega_s and omega_a are variables of the theorem, and in experiments specific points inside the monotonicity region M are chosen. No invented entities appear.

free parameters (1)
  • D1Q3 link coefficient L2 = 12/25 or 1/3 in numerical experiments
    The coefficient L2 in the equilibrium split (3.1) is a free design parameter of the scheme, chosen by hand in the tests. The theorem holds for any L_i satisfying (3.4); it is not fitted to data.
assumptions (4)
  • ad hoc to paper Equilibria split as f_eq_i(u) = L_i u + sum_{k=1..d} N_{i,k} phi_k(u), with symmetry L_{2l}=L_{2l+1}, N_{2l,k}=-N_{2l+1,k} >= 0.
    Equations (3.1) and (3.3). This structural restriction is load-bearing: all Jacobian computations in Prop. 3.3 and the link-wise decomposition in Prop. 4.4 rely on it. It is standard in kinetic BGK models but not universal for TRT equilibria.
  • domain assumption Fluxes phi_k are C^1 with phi_k(0)=0 and initial data u0 in L1(Rd) intersect L-infinity(Rd) intersect BV(Rd).
    Section 1. Standard well-posedness framework for scalar conservation laws; also gives u_infinity = ||u0||_infinity for the invariant interval.
  • domain assumption Initialization of the scheme at equilibrium, f^0_i,j = f_eq_i(volume average of u0 on cell C_j), equation (2.2).
    Used in Prop. 4.5 to identify the L1 distance between two solutions at time zero with the L1 distance between initial data, and in the invariant compact set induction. The paper notes this is standard for first-order schemes.
  • standard math Crandall-Majda compactness and Lax-Wendroff-type consistency machinery for passing from numerical estimates to a weak solution.
    Invoked in Thm. 4.10 via [12] and in the flux-term limit; assumed as background.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Monotonicity and convergence of two-relaxation-times lattice Boltzmann schemes for a non-linear conservation law." pith.science (2026). https://pith.science/paper/CK6R45H7

@misc{pith2026250107934,
  author       = {Pith},
  title        = {Pith review of: Monotonicity and convergence of two-relaxation-times lattice Boltzmann schemes for a non-linear conservation law},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CK6R45H7}},
  note         = {Machine review of arXiv:2501.07934}
}
abstract

We address the convergence analysis of lattice Boltzmann methods for scalar non-linear conservation laws, focusing on two-relaxation-times (TRT) schemes. Unlike Finite Difference/Finite Volume methods, lattice Boltzmann schemes offer exceptional computational efficiency and parallelization capabilities. However, their monotonicity and $L^{\infty}$-stability remain underexplored. Extending existing results on simpler BGK schemes, we derive conditions ensuring that TRT schemes are monotone and stable by leveraging their unique relaxation structure. Our analysis culminates in proving convergence of the numerical solution to the weak entropy solution of the conservation law. Compared to BGK schemes, TRT schemes achieve reduced numerical diffusion while retaining provable convergence. Numerical experiments validate and illustrate the theoretical findings.

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. Equilibrium boundary conditions for vectorial multi-dimensional lattice Boltzmann schemes

    math.NA 2025-05 conditional novelty 7.0 of 10

    Equilibrium boundary conditions for monotone lattice Boltzmann schemes are shown to converge to the Bardos-Leroux-Nedelec entropy solution in the scalar multidimensional case.

Reference graph

Works this paper leans on

28 extracted references · 28 canonical work pages · cited by 1 Pith paper

  1. [1]

    D. Aregba-Driollet , Convergence of Lattice Boltzmann methods with overrelaxation for a nonlinear conservation law , ESAIM: Mathematical Modelling and Numerical Analysis, 58 (2024), pp. 1935–1958

    work page 2024

  2. [2]

    Aregba-Driollet and R

    D. Aregba-Driollet and R. Natalini, Discrete kinetic schemes for multidimensional systems of conservation laws , SIAM Journal on Numerical Analysis, 37 (2000), pp. 1973–2004

    work page 2000

  3. [3]

    T. Bellotti, Monotonicity for genuinely multi-step methods: Results and issues from a simple lattice Boltzmann scheme , in International Conference on Finite Volumes for Complex Applications, Springer, 2023, pp. 33–41

    work page 2023

  4. [4]

    Bellotti , Numerical analysis of lattice Boltzmann schemes: from fundamental issues to efficient and accurate adaptive methods, PhD thesis, Institut Polytechnique de Paris, 2023

    T. Bellotti , Numerical analysis of lattice Boltzmann schemes: from fundamental issues to efficient and accurate adaptive methods, PhD thesis, Institut Polytechnique de Paris, 2023

    work page 2023

  5. [5]

    T. Bellotti, Truncation errors and modified equations for the lattice Boltzmann method via the corresponding Finite Difference schemes , ESAIM: Mathematical Modelling and Nu- merical Analysis, 57 (2023), pp. 1225–1255

    work page 2023

  6. [6]

    T. Bellotti , Initialisation from lattice Boltzmann to multi-step Finite Difference methods: modified equations and discrete observability , Journal of Computational Physics, 504 (2024), p. 112871

    work page 2024

  7. [7]

    Bellotti, B

    T. Bellotti, B. Graille, and M. Massot, Finite Difference formulation of any lattice Boltz- mann scheme , Numerische Mathematik, 152 (2022), pp. 1–40

    work page 2022

  8. [8]

    Bouchut, Construction of BGK models with a family of kinetic entropies for a given system of conservation laws , Journal of Statistical Physics, 95 (1999), pp

    F. Bouchut, Construction of BGK models with a family of kinetic entropies for a given system of conservation laws , Journal of Statistical Physics, 95 (1999), pp. 113–170

    work page 1999

Show all 28 references
  1. [9]

    Brenner, V

    P. Brenner, V. Thom ´ee, and L. B. W ahlbin , Besov spaces and applications to difference methods for initial value problems , vol. 434, Springer, 2006

    2006

  2. [10]

    Caetano, F

    F. Caetano, F. Dubois, and B. Graille , A result of convergence for a mono-dimensional two-velocities lattice boltzmann scheme , Discrete and Continuous Dynamical Systems - S, 17 (2024), pp. 3129–3154, https://doi.org/10.3934/dcdss.2023072

    2024 doi

  3. [11]

    Y. Chen, Z. Chai, and B. Shi , A unified fourth-order Bhatnagar-Gross-Krook lat- tice Boltzmann model for high-dimensional linear hyperbolic equations , arXiv preprint arXiv:2410.13165, (2024)

    2024 arXiv

  4. [12]

    M. G. Crandall and A. Majda, Monotone difference approximations for scalar conservation laws, Mathematics of Computation, 34 (1980), pp. 1–21

    1980

  5. [13]

    Dellacherie , Construction and analysis of lattice Boltzmann methods applied to a 1D convection-diffusion equation, Acta Applicandae Mathematicae, 131 (2014), pp

    S. Dellacherie , Construction and analysis of lattice Boltzmann methods applied to a 1D convection-diffusion equation, Acta Applicandae Mathematicae, 131 (2014), pp. 69–140

    2014

  6. [14]

    P. J. Dellar , A magic two-relaxation-time lattice Boltzmann algorithm for magnetohydrody- namics, Discrete and Continuous Dynamical Systems-S, 17 (2024), pp. 3155–3173

    2024

  7. [15]

    d’Humi`eres, Generalized lattice-Boltzmann equations, Rarefied gas dynamics, (1992)

    D. d’Humi`eres, Generalized lattice-Boltzmann equations, Rarefied gas dynamics, (1992)

    1992

  8. [16]

    Dubois, Nonlinear fourth order Taylor expansion of lattice Boltzmann schemes, Asymptotic Analysis, 127 (2022), pp

    F. Dubois, Nonlinear fourth order Taylor expansion of lattice Boltzmann schemes, Asymptotic Analysis, 127 (2022), pp. 297–337

    2022

  9. [17]

    Dubois, B

    F. Dubois, B. Graille, and S. R. Rao, A notion of non-negativity preserving relaxation for a mono-dimensional three velocities scheme with relative velocity , Journal of Computational Science, 47 (2020), p. 101181

    2020

  10. [18]

    Ginzburg, Variably saturated flow described with the anisotropic lattice Boltzmann methods, Computers & Fluids, 35 (2006), pp

    I. Ginzburg, Variably saturated flow described with the anisotropic lattice Boltzmann methods, Computers & Fluids, 35 (2006), pp. 831–848. 26 D. AREGBA-DRIOLLET, AND T. BELLOTTI

    2006

  11. [19]

    I. Ginzburg , Une variation sur les propri´ et´ es magiques de mod` eles de Boltzmann pour l’´ ecoulement microscopique et macroscopique, Th` ese d’Habilitation ` a diriger des recherches, Universit´ e Pierre et Marie Curie Paris, 2009

    2009

  12. [20]

    Ginzburg, F

    I. Ginzburg, F. Verhaeghe, and D. d’Humi `eres, Two-relaxation-time lattice Boltzmann scheme: About parametrization, velocity, pressure and mixed boundary conditions , Com- munications in Computational Physics, 3 (2008), pp. 427–478

    2008

  13. [21]

    Godlewski and P.-A

    E. Godlewski and P.-A. Raviart, Hyperbolic systems of conservation laws , no. 3-4, Ellipses, 1991

    1991

  14. [22]

    Graille , Approximation of mono-dimensional hyperbolic systems: A lattice Boltzmann scheme as a relaxation method , Journal of Computational Physics, 266 (2014), pp

    B. Graille , Approximation of mono-dimensional hyperbolic systems: A lattice Boltzmann scheme as a relaxation method , Journal of Computational Physics, 266 (2014), pp. 74–88

    2014

  15. [23]

    Hundsdorfer and S

    W. Hundsdorfer and S. Ruuth , On monotonicity and boundedness properties of linear mul- tistep methods, Mathematics of Computation, 75 (2006), pp. 655–672

    2006

  16. [24]

    Hundsdorfer, S

    W. Hundsdorfer, S. J. Ruuth, and R. J. Spiteri , Monotonicity-preserving linear multistep methods, SIAM Journal on Numerical Analysis, 41 (2003), pp. 605–623

    2003

  17. [25]

    Natalini, A discrete kinetic approximation of entropy solutions to multidimensional scalar conservation laws, Journal of Differential Equations, 148 (1998), pp

    R. Natalini, A discrete kinetic approximation of entropy solutions to multidimensional scalar conservation laws, Journal of Differential Equations, 148 (1998), pp. 292–317

    1998

  18. [26]

    Sabac, The optimal convergence rate of monotone finite difference methods for hyperbolic conservation laws, SIAM Journal on Numerical Analysis, 34 (1997), pp

    F. Sabac, The optimal convergence rate of monotone finite difference methods for hyperbolic conservation laws, SIAM Journal on Numerical Analysis, 34 (1997), pp. 2306–2318

    1997

  19. [27]

    Teng, Error bound between monotone difference schemes and their modified equations , Mathematics of Computation, 79 (2010), pp

    Z.-H. Teng, Error bound between monotone difference schemes and their modified equations , Mathematics of Computation, 79 (2010), pp. 1473–1491

    2010

  20. [28]

    R. F. W arming and B. J. Hyett, The modified equation approach to the stability and accuracy analysis of finite-difference methods, Journal of Computational Physics, 14 (1974), pp. 159– 179

    1974

Pith tools

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