Pith. sign in

REVIEW 2 major objections 5 minor 35 references

A structure-preserving local discontinuous Galerkin method for the Fokker-Planck-Landau equation

T0 review · 2 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read This paper claims that an LDG discretization of the Fokker-Planck-Landau equation can conserve mass, momentum, and energy while adding upwind stabilization, by preserving the collision kernel's null-space structure at the discrete level.

desk verdict A genuine but incremental extension of Shiroto-Sentoku to LDG; the conservation mechanism is sound, but the advertised L2 stability identity has a boundary-term gap and the numerics are too thin to close it. read the letter →

arxiv 2505.18321 v1 pith:5C57XPJX submitted 2025-05-23 math.NA cs.NA

Kun Huang , Andrés Galindo-Olarte , Rodrigo González-Hernández , Irene M. Gamba This is my paper · ORCID
classification math.NAcs.NA MSC 65M6065N3035Q8382C40
keywords Fokker-Planck-LandauequationlocaldiscontinuousGalerkinstructure-preservingmethoddiscreteconservationlawsupwindfluxgradientoperatorcollisionkernelnullspacerelativisticLandau
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 claims that a local discontinuous Galerkin discretization of the Fokker-Planck-Landau equation can simultaneously conserve mass, momentum, and energy and add upwind stabilization, a combination previously seen as contradictory. The key is to design the discrete collision kernel through a projection of the particle kinetic energy so that the continuous null-space identity—relative velocity lies in the kernel's null space—survives at the discrete level, while the jump terms of the LDG framework absorb the upwind dissipation. If correct, the scheme would give high-order, parallelizable, entropy-dissipative simulations of collisional plasmas without sacrificing the collision operator's geometric structure.

What carries the argument

The central object is the discrete gradient operator ∇_h, defined via the alternating flux in (22)–(23), which acts as a projection of the gradient and equals the interior gradient plus a lift of interface jumps. The structure-preserving collision kernel is Φ_h(p,q)=Λ(p,q) S(∇_{h,p}Π_h E(p), ∇_{h,q}Π_h E(q)), so the continuous null-space identity S(v,w)·(v−w)=0 carries over to the discrete level. Because Π_h E is chosen in $C^{1}$, the advection field U_h is continuous and the upwind flux is well-defined; the strictly dissipative term in (39) is a jump term that vanishes when the test function is one of the conserved quantities.

What would settle it

Run the scheme with k=1 elements for the non-relativistic equation and monitor momentum and energy; any drift beyond round-off would falsify the $C^{1}$ projection assumption, and evaluating both sides of (39) on a mesh with non-empty boundary would reveal whether the boundary treatment is consistent.

Watch

Extended reading notes

Core claim

The paper's central claim is that the upwind structure-preserving LDG scheme (37) satisfies discrete conservation of mass, momentum, and projected kinetic energy, and an L2 stability identity with a dissipative jump term, provided a projection Π_h into W_h∩$C^{1}$(Ω) that fixes the conserved quantities exists. For the non-relativistic equation with E(p)=|p|^2/2, this holds for polynomial degree at least 2; for the relativistic equation E(p)=$\sqrt$(1+|p|^2), only the projected energy is conserved. The conservation follows from the identity S(v,w)·(v−w)=0 for the collision tensor, which the discrete kernel Φ_h inherits by construction, and the upwind stabilization enters through jump terms that vanish for the conserved test functions.

Load-bearing premise

The proof hinges on a projection of the kinetic energy into the space of continuous piecewise polynomials of degree at least 2, and on the boundary terms in the discrete forms cancelling consistently; if either fails, conservation or the L2 stability identity does not follow.

Editorial extensions

If this is right

  • On uniform Cartesian meshes with Q2 elements, the scheme conserves mass, momentum, and energy to machine precision while entropy decays toward the Maxwellian, as shown in the numerical experiments.
  • The upwind flux adds L2 dissipation without breaking conservation, because the dissipative contribution is a sum over element interfaces of ⟨|U_h·n|/2⟩[f_h],[f_h]⟩ terms that vanish for the conserved test functions.
  • The method reproduces the Shiroto-Sentoku finite-difference scheme when the basis is piecewise constant, and the conservative finite-element scheme of Hirvijoki when elements are continuous.
  • High-order accuracy and parallelization are available because the discrete problem can be advanced element by element after evaluating the nonlocal fields D_h and U_h.
  • The framework applies to both non-relativistic and relativistic FPL equations with any interaction potential admitted by the tensor form.

Reading between the lines

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

  • A natural test is to lower the polynomial degree to k=1 in the non-relativistic case, where Π_h E no longer lies in C^1; if conservation persists numerically, the C^1 requirement is stronger than needed.
  • The same recipe—replace the kernel's energy arguments by projected quantities so that the null-space identity holds discretely—could be applied to other collision models with tensor kernels, such as the quasilinear diffusion equation mentioned in the paper.
  • For the relativistic equation, the scheme conserves a projected energy rather than the physical energy; whether the difference stays small over long times in runaway-electron simulations is an open question.
  • The scheme is not positivity-preserving, so combining it with a positivity limiter would show whether the conservation and entropy properties survive the correction.
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

2 major / 5 minor

Summary. The paper proposes a local discontinuous Galerkin (LDG) method for the Fokker-Planck-Landau equation that is designed to preserve mass, momentum, and (projected) energy while adding upwind stabilization. The key idea is to replace the continuous collision kernel by a discrete kernel built from the discrete gradient of a projected kinetic energy, so that the null-space property S(v,w)\cdot(v-w)=0 is preserved at the discrete level. A symmetric structure-preserving scheme (Theorem 3) and an upwind structure-preserving scheme (Theorem 4) are stated, together with an L2 identity containing a dissipative jump term. Numerical experiments are reported for a non-relativistic Q^2 approximation on an 8^3 mesh, showing conservation to machine precision and decay of a relative-entropy diagnostic.

Significance. If the theorems are made fully correct, the paper gives an elegant resolution of the usual tension between exact discrete conservation and upwind stabilization: conservation is tested against continuous functions in W_h\cap C^1(\Omega), while stability is obtained by testing against discontinuous f_h so that jump terms contribute dissipation. The construction of the discrete kernel through discrete gradients of a projected energy is a clean rephrasing of the Shiroto-Sentoku strategy in the LDG context and is largely basis-agnostic. The numerical example supports the main qualitative claims. The main limitations are that there is no convergence or error analysis, no positivity preservation, and for the relativistic case only a projected energy is conserved rather than the physical energy; these are stated or implicit in the paper.

major comments (2)
  1. [§4.2, Theorem 3, Eq. (30)-(32)] The proof that momentum conservation is 'trivial' is not valid for the stated level of generality. For \phi_h=\Pi_h p_i, the right-hand side of (30) vanishes only if \nabla_{h,q}\Pi_h p_i(q)-\nabla_{h,p}\Pi_h p_i(p)=0, which requires that the discrete gradient of the projected coordinate be a constant field. This holds when p_i\in W_h and \Pi_h fixes W_h (in particular for Q^k with k\ge 1), and it may hold on uniform meshes for piecewise constants, but it fails for arbitrary projections on nonuniform meshes, where \nabla_h(\Pi_h p_i) varies from cell to cell. The theorem should either state the additional hypothesis or prove the required discrete-gradient property; as written, the claimed basis-independence of the structure-preserving strategy is not established.
  2. [§4.3, Theorem 4, Eq. (37)-(39)] The derivation of the L2 identity (39) from the upwind scheme (37) is compressed into the sentence 'Substitute \varphi_h with f_h', but the appearance of the boundary term -\langle (1/2)U_h\cdot n^-, (f_h^-)^2\rangle_\Gamma is nontrivial because (37) contains no numerical flux on \Gamma. The authors should expand the proof by applying the divergence theorem to \sum_R (U_h f_h,\nabla f_h)_R, combining interior face contributions with the upwind flux (41), and explicitly stating the discrete boundary treatment on \Gamma. With that derivation, (39) is consistent, but the statement that 'L2 stability is enhanced' should also be qualified: (39) is an identity with an indefinite boundary contribution and an indefinite (1/2)\nabla\cdot U_h term, not a bound on \|f_h\| alone.
minor comments (5)
  1. [§3.2, Eq. (25)] There appears to be a sign error in the displayed formula: the identity \sum_R(\nabla_h f_h,V_h)_R = \sum_R(\nabla f_h,V_h)_R + \sum_R\langle f_h^- n^-, \{V_h\}\rangle_{\partial R\setminus\Gamma} conflicts with the proof around Eq. (28) and with the lift definition, which imply a minus sign in front of the interface term.
  2. [§3.1, Eq. (10) and §4.3, Eq. (37)] The zero-flux Robin condition (10) is stated for the continuous problem, but the discrete scheme (37) imposes no numerical flux on \Gamma while also omitting boundary flux sums. The authors should state explicitly which discrete boundary flux is used on \Gamma; this will make the derivation of (39) unambiguous.
  3. [§4.3, after Eq. (40)] The continuity of U_h follows only if the scalar factor \Lambda(p,q) is regular enough on the diagonal p=q; for Coulomb-type kernels with singular \Lambda the composition argument needs additional assumptions. Please state the required regularity on \Lambda.
  4. [§6.2, entropy diagnostic] The plotted quantity H(f|M)=\int f^+\ln f^+ - \int M\ln M is the relative entropy only when f^+ has the same mass, momentum, and energy as M; since f is not positivity-preserving, the relation to the true relative entropy should be clarified or the diagnostic corrected.
  5. [Throughout] There are several typographical issues, including 'discontinous' in the abstract, inconsistent spacing in sums, and the undefined notation \widehat{U_h f_h} in (37) before its definition in (41); please proofread and ensure all flux notation is defined before use.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the conservation laws are consequences of an explicitly designed discrete kernel, and no fitted parameter is relabeled as a prediction.

full rationale

The central conservation claims are design theorems rather than disguised inputs. In Section 4.1 the paper states that 'a structure-preserving scheme can be obtained by replacing the collision kernel Φ with a specifically designed approximation Φ_h,' and Theorem 3 defines Φ_h in equation (31) through S(∇_{h,p}Π_h E, ∇_{h,q}Π_h E). Energy conservation then follows from the algebraic null-space identity S(v,w)·(v−w)=0 quoted in equation (33), so the result is built into the construction; this is standard structure preservation, not circular prediction. Mass and momentum conservation are the same explicit algebraic mechanism. Theorem 4's stability identity is obtained by substituting φ_h=f_h into (37), and no fitted value or external benchmark is reverse-engineered anywhere in the derivation. The skeptic's boundary-term concern about (39) versus (37) is a proof-consistency question, not a circularity, because the identity is not assumed as an input. Self-citations such as [22], [25-27], and [34-35] are either contextual references or an analogy for temporal discretization choices; none of them supplies an unverified uniqueness theorem or carries the load of the main argument. Therefore there is no significant circularity.

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

No free parameters are fitted to data. The central construction adds no new physical entities; the discrete kernel Phi_h and the discrete gradient operator are mathematical devices. The main assumed ingredients are the Gaussian fall-off of the continuum solution and the existence of a C^1 projection of the energy into the DG space.

assumptions (4)
  • domain assumption The continuum FPL solution is bounded by a Gaussian determined by the initial data, so a finite computational domain with homogeneous boundary conditions is valid.
    Used in Section 3.1 to justify truncating R^3 to a bounded box; relies on [18] and excludes boundary effects in conservation statements.
  • standard math The discrete gradient operator satisfies nabla_h phi_h = nabla phi_h + lift(rr phi_h ss) under the alternating fluxes (22)-(23).
    Proposition 1 and Equation (26), a standard LDG result from [1]; the proof checks out.
  • standard math The collision tensor S(v,w) satisfies S(v,w) dot (v - w) = 0 and S is positive semidefinite.
    Section 2; this is the algebraic null-space property that the scheme is designed to reproduce.
  • domain assumption There exists a projection Pi_h into W_h intersected with C^1(Omega) that fixes 1, p and E(p).
    Theorem 4 and Remark 3; for non-relativistic FPL with k at least 2 this is satisfied by the L2 projection, but for relativistic FPL only a projected energy is fixed.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A structure-preserving local discontinuous Galerkin method for the Fokker-Planck-Landau equation." pith.science (2026). https://pith.science/paper/5C57XPJX

@misc{pith2026250518321,
  author       = {Pith},
  title        = {Pith review of: A structure-preserving local discontinuous Galerkin method for the Fokker-Planck-Landau equation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5C57XPJX}},
  note         = {Machine review of arXiv:2505.18321}
}
read the original abstract

In this work, we introduce a structure-preserving local discontinuous Galerkin (LDG) method \cite{cockburn1998local} for solving the non-local non-linear Fokker-Planck-Landau (FPL) equations. We rephrase the structure-preserving strategy of Shiroto and Sentoku\cite{shiroto2019structure} in the language of numerical analysis, and extend it to the LDG framework. We propose a method that is not only conservative, but also stabilized through upwind flux. The apparent contradiction between conservation laws and numerical stabilization is elegantly resolved by leveraging the properties of the jump terms inherent to the LDG framework. In the numerical experiments, our scheme is tested with benchmark examples.

Figures

Figures reproduced from arXiv: 2505.18321 by the authors.

Figure 1
Figure 1. Cross section plot of fpp, tq at pz “ 0 in different time. The simulation was performed on a 8 ˆ 8 ˆ 8 mesh with Q2 basis functions. 13 [PITH_FULL_IMAGE:figures/full_fig_p013_1.png] view at source ↗
Figure 2
Figure 2. Conservation errors and the decay of entropy. The simulation was performed on a 8 [PITH_FULL_IMAGE:figures/full_fig_p014_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 31 canonical work pages

  1. [1]

    Unified analy- sis of discontinuous galerkin methods for elliptic problems.SIAM journal on numerical analysis, 39(5):1749–1779, 2002

    Douglas N Arnold, Franco Brezzi, Bernardo Cockburn, and L Donatella Marini. Unified analy- sis of discontinuous galerkin methods for elliptic problems.SIAM journal on numerical analysis, 39(5):1749–1779, 2002

    work page 2002

  2. [2]

    S. T. Beliaev and G. I. Budker. The Relativistic Kinetic Equation.Soviet Physics Doklady, 1:218, October 1956

    work page 1956

  3. [3]

    The relativistic kinetic equation

    ST Beliaev and GI Budker. The relativistic kinetic equation. InSoviet physics doklady, volume 1, page 218, 1956

    work page 1956

  4. [4]

    Boltzmann’s equation for electron gas in which collisions are infrequent

    ST Belyaev and GI Budker. Boltzmann’s equation for electron gas in which collisions are infrequent. Plasma Physics and the Problem of Controlled Thermonuclear Reactions, edited by MA Leontovich (Pergamon Press, London, 1959), 2:431, 1961

    work page 1959

  5. [5]

    Fast algorithms for numerical, conservative, and entropy approximations of the fokker–planck–landau equation.Journal of Com- putational Physics, 133(2):310–322, 1997

    C Buet, St´ ephane Cordier, Pierre Degond, and Mohamed Lemou. Fast algorithms for numerical, conservative, and entropy approximations of the fokker–planck–landau equation.Journal of Com- putational Physics, 133(2):310–322, 1997

    work page 1997

  6. [6]

    A particle method for the homogeneous landau equation.Journal of Computational Physics: X, 7:100066, 2020

    Jose A Carrillo, Jingwei Hu, Li Wang, and Jeremy Wu. A particle method for the homogeneous landau equation.Journal of Computational Physics: X, 7:100066, 2020

    2020

  7. [7]

    An a priori error analysis of the local discontinuous galerkin method for elliptic problems.SIAM Journal on Numerical Analysis, 38(5):1676–1706, 2000

    Paul Castillo, Bernardo Cockburn, Ilaria Perugia, and Dominik Sch¨ otzau. An a priori error analysis of the local discontinuous galerkin method for elliptic problems.SIAM Journal on Numerical Analysis, 38(5):1676–1706, 2000

    work page 2000

  8. [8]

    Some extensions of the local discontinuous galerkin method for convection-diffusion equations in multidimensions.Mathematics of Finite Elements and Appli- cations, 10:225–238, 1999

    Bernardo Cockburn and Clint Dawson. Some extensions of the local discontinuous galerkin method for convection-diffusion equations in multidimensions.Mathematics of Finite Elements and Appli- cations, 10:225–238, 1999. 15

    work page 1999

Show all 35 references
  1. [9]

    Superconvergence of the local discontinuous galerkin method for elliptic problems on cartesian grids.SIAM Journal on Numerical Analysis, 39(1):264–285, 2001

    Bernardo Cockburn, Guido Kanschat, Ilaria Perugia, and Dominik Sch¨ otzau. Superconvergence of the local discontinuous galerkin method for elliptic problems on cartesian grids.SIAM Journal on Numerical Analysis, 39(1):264–285, 2001

    2001

  2. [10]

    The local discontinuous galerkin method for time-dependent convection-diffusion systems.SIAM journal on numerical analysis, 35(6):2440–2463, 1998

    Bernardo Cockburn and Chi-Wang Shu. The local discontinuous galerkin method for time-dependent convection-diffusion systems.SIAM journal on numerical analysis, 35(6):2440–2463, 1998

    1998

  3. [11]

    Runge–kutta discontinuous galerkin methods for convection- dominated problems.Journal of scientific computing, 16:173–261, 2001

    Bernardo Cockburn and Chi-Wang Shu. Runge–kutta discontinuous galerkin methods for convection- dominated problems.Journal of scientific computing, 16:173–261, 2001

    2001

  4. [12]

    A fully implicit, scalable, conservative nonlinear relativistic fokker–planck 0d-2p solver for runaway electrons.Computer Physics Communications, 254:107361, 2020

    Don Daniel, William T Taitano, and Luis Chac´ on. A fully implicit, scalable, conservative nonlinear relativistic fokker–planck 0d-2p solver for runaway electrons.Computer Physics Communications, 254:107361, 2020

    2020

  5. [13]

    An entropy scheme for the fokker-planck collision operator of plasma kinetic theory.Numerische Mathematik, 68(2):239–262, 1994

    Pierre Degond and Brigitte Lucquin-Desreux. An entropy scheme for the fokker-planck collision operator of plasma kinetic theory.Numerische Mathematik, 68(2):239–262, 1994

    1994

  6. [14]

    Numerical methods for plasma physics in collisional regimes.Journal of Plasma Physics, 81(1):305810106, 2015

    Giacomo Dimarco, Qin Li, Lorenzo Pareschi, and Bokai Yan. Numerical methods for plasma physics in collisional regimes.Journal of Plasma Physics, 81(1):305810106, 2015

    2015

  7. [15]

    Numerical methods for kinetic equations.Acta Numerica, 23:369–520, 2014

    Giacomo Dimarco and Lorenzo Pareschi. Numerical methods for kinetic equations.Acta Numerica, 23:369–520, 2014

    2014

  8. [16]

    Model fokker-planck equation for a plasma and its solution.The Physics of Fluids, 7(11):1788–1799, 1964

    JP Dougherty. Model fokker-planck equation for a plasma and its solution.The Physics of Fluids, 7(11):1788–1799, 1964

    1964

  9. [17]

    Electron and ion runaway in a fully ionized gas

    Harry Dreicer. Electron and ion runaway in a fully ionized gas. i.Physical Review, 115(2):238, 1959

    1959

  10. [18]

    The landau equation does not blow up.arXiv preprint arXiv:2311.09420, 2023

    Nestor Guillen and Luis Silvestre. The landau equation does not blow up.arXiv preprint arXiv:2311.09420, 2023

    2023 arXiv

  11. [19]

    Collisional transport in plasma.Handbook of Plasma Physics, 1(147):331, 1983

    Fred L Hinton. Collisional transport in plasma.Handbook of Plasma Physics, 1(147):331, 1983

    1983

  12. [20]

    Conservative finite-element method for the relativistic coulomb collision operator

    Eero Hirvijoki. Conservative finite-element method for the relativistic coulomb collision operator. arXiv preprint arXiv:1903.07403, 2019

    1903 arXiv

  13. [21]

    Conservative discretization of the landau collision integral

    Eero Hirvijoki and Mark F Adams. Conservative discretization of the landau collision integral. Physics of Plasmas, 24(3), 2017

    2017

  14. [22]

    A conservative galerkin solver for the quasilinear diffusion model in magnetized plasmas.Journal of Computational Physics, 488:112220, 2023

    Kun Huang, Michael Abdelmalik, Boris Breizman, and Irene M Gamba. A conservative galerkin solver for the quasilinear diffusion model in magnetized plasmas.Journal of Computational Physics, 488:112220, 2023

    2023

  15. [23]

    Nonlinear fokker- planck collision operator in rosenbluth form for gyrokinetic simulations using discontinuous galerkin method.Computer Physics Communications, 279:108459, 2022

    Dongkyu Kim, Janghoon Seo, Gahyung Jo, Jae-Min Kwon, and Eisung Yoon. Nonlinear fokker- planck collision operator in rosenbluth form for gyrokinetic simulations using discontinuous galerkin method.Computer Physics Communications, 279:108459, 2022

    2022

  16. [24]

    Pergamon Press, 1981

    Evgenii Mikhailovich Lifshitz and Lev Petrovich Pitaevskii.Course of theoretical physics. Pergamon Press, 1981

    1981

  17. [25]

    Decay of entropy from a conservative spectral method for fokker-planck-landau type equations

    Clark A Pennie and Irene M Gamba. Decay of entropy from a conservative spectral method for fokker-planck-landau type equations. InAIP Conference Proceedings, volume 2132. AIP Publishing, 2019

    2019

  18. [26]

    Entropy decay rates for conservative spectral schemes modeling fokker-planck-landau type flows in the mean field limit.arXiv preprint arXiv:1910.03110, 2019

    Clark A Pennie and Irene M Gamba. Entropy decay rates for conservative spectral schemes modeling fokker-planck-landau type flows in the mean field limit.arXiv preprint arXiv:1910.03110, 2019

    1910 arXiv

  19. [27]

    Convergence and error estimates for the conservative spectral method for fokker-planck-landau equations.arXiv preprint arXiv:2009.10352, 2020

    Clark A Pennie and Irene M Gamba. Convergence and error estimates for the conservative spectral method for fokker-planck-landau equations.arXiv preprint arXiv:2009.10352, 2020

    2009 arXiv

  20. [28]

    A mass-energy- conserving discontinuous galerkin scheme for the isotropic multispecies rosenbluth–fokker–planck equation.Journal of Computational Physics, 449:110813, 2022

    Takashi Shiroto, Akinobu Matsuyama, Nobuyuki Aiba, and Masatoshi Yagi. A mass-energy- conserving discontinuous galerkin scheme for the isotropic multispecies rosenbluth–fokker–planck equation.Journal of Computational Physics, 449:110813, 2022. 16

    2022

  21. [29]

    Structure-preserving strategy for conservative simulation of the relativistic nonlinear landau-fokker-planck equation.Physical Review E, 99(5):053309, 2019

    Takashi Shiroto and Yasuhiko Sentoku. Structure-preserving strategy for conservative simulation of the relativistic nonlinear landau-fokker-planck equation.Physical Review E, 99(5):053309, 2019

    2019

  22. [30]

    Entropy dissipation estimates for the relativistic landau equa- tion, and applications.Journal of Functional Analysis, 277(4):1139–1201, 2019

    Robert M Strain and Maja Taskovi´ c. Entropy dissipation estimates for the relativistic landau equa- tion, and applications.Journal of Functional Analysis, 277(4):1139–1201, 2019

    2019

  23. [31]

    Variational crimes in the finite element method

    Gilbert Strang. Variational crimes in the finite element method. InThe mathematical foundations of the finite element method with applications to partial differential equations, pages 689–710. Elsevier, 1972

    1972

  24. [32]

    William T Taitano, Luis Chac´ on, AN Simakov, and K Molvig. A mass, momentum, and energy conserving, fully implicit, scalable algorithm for the multi-dimensional, multi-species rosenbluth– fokker–planck equation.Journal of Computational Physics, 297:357–380, 2015

    2015

  25. [33]

    Springer Berlin Heidelberg New York, 1996

    Gerhard Wanner and Ernst Hairer.Solving ordinary differential equations II, volume 375. Springer Berlin Heidelberg New York, 1996

    1996

  26. [34]

    A conservative scheme for vlasov poisson landau modeling collisional plasmas.Journal of Computational Physics, 340:470–497, 2017

    Chenglong Zhang and Irene M Gamba. A conservative scheme for vlasov poisson landau modeling collisional plasmas.Journal of Computational Physics, 340:470–497, 2017

    2017

  27. [35]

    A conservative discontinuous galerkin solver for the space homogeneous boltzmann equation for binary interactions.SIAM Journal on Numerical Analysis, 56(5):3040–3070, 2018

    Chenglong Zhang and Irene M Gamba. A conservative discontinuous galerkin solver for the space homogeneous boltzmann equation for binary interactions.SIAM Journal on Numerical Analysis, 56(5):3040–3070, 2018. 17

    2018

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.