Pith. sign in

REVIEW 3 major objections 4 minor 33 references

An L-Stable Sequential Two-Stage Fourth-Order Method with ADER Trajectory Derivatives for Stiff Transport--Relaxation Systems

T0 review · 3 major / 4 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read Two-stage fourth-order implicit integrator for stiff transport–relaxation equations proves L-stability and asymptotic preservation by pairing a finite-volume residual with its trajectory derivative.

desk verdict A careful, honestly-scoped paper; the new ADER closure, C_q parameter, and AP theory are solid for the linear/exact-closure case, but nonlinear ADER closure remains unverified. read the letter →

arxiv 2608.03256 v1 pith:NF47CGAE submitted 2026-08-04 math.NA cs.NA

Zhixin Huo , Yangnan Su This is my paper · ORCID
classification math.NAcs.NA MSC 65L0665M0865M12
keywords two-derivativetimeintegrationL-stabilitystifftransport–relaxationsystemsADERtrajectoryderivativeconservativefinite-volumemethodasymptoticpreservationuniformaccuracysequentialimplicitstages
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 tries to establish that a fully implicit, two-stage, fourth-order time integrator can be closed for stiff transport–relaxation equations as a complete unsplit method, not just a temporal formula. It supplies the missing object—the discrete trajectory derivative $\mathcal G_h^{\mathrm{tr}}=\mathcal D\mathcal L_h\,\mathcal L_h$ of a conservative finite-volume residual—and approximates it at faces from the derivative of the numerical flux, fed by local ADER/Cauchy–Kowalevski predictors. In the linear constant-coefficient case the face-based approximation is exactly $\mathcal L_h^2$, even though it is assembled independently, so the fourth-order temporal theory transfers unchanged. The paper proves the completed step is fourth order and L-stable, that the parameter $C_q=5/183$ changes the deep-stiff amplification from $O(|z|^{-1})$ to $O(|z|^{-2})$, and that for fixed compatible spatial operators the full step reaches the discrete diffusion limit with error $O(\delta)$. If these claims hold, one method damps unresolved fast waves, keeps fourth-order accuracy, and passes automatically to the relaxation limit without splitting transport from the stiff source.

What carries the argument

The central object is the discrete trajectory derivative $\mathcal G_h^{\mathrm{tr}}(Z)=\mathcal D\mathcal L_h(Z)\,\mathcal L_h(Z)$, the time derivative of the semi-discrete conservative residual along its own trajectory. Each implicit stage is closed with this derivative, supplied either exactly or through the ADER/Cauchy–Kowalevski flux chain rule $\widehat{\mathbf F}_t=\widehat{\mathbf F}_{U^-}U^-_t+\widehat{\mathbf F}_{U^+}U^+_t$ with shared face values, so conservation is preserved by construction. In the linear constant-coefficient case this produces the exact identity $\widetilde{\mathcal G}_h=\mathcal G_h^{\mathrm{tr}}=\mathcal L_h^2$ without forming the square of the residual matrix

What would settle it

Run a smooth nonlinear relaxation problem with a quadratic flux on a fixed fine grid and close the two stages with the ADER face-based $\widetilde{\mathcal G}_h$ only, never the exact $\mathcal G_h^{\mathrm{tr}}$. If the measured temporal order drops below four, or if $\|\widetilde{\mathcal G}_h(\Pi_h U)-\mathcal G_h^{\mathrm{tr}}(\Pi_h U)\|$ does not decay like $O(h^p)$, then the trajectory-closure bound fails and the fourth-order claim for the ADER-implemented nonlinear method is not in force.

Watch

Extended reading notes

Core claim

The central claim is that the sequential two-stage two-derivative step, closed with the conservative spatial pair $(\mathcal L_h,\mathcal G_h^{\mathrm{tr}})$, delivers simultaneous fourth-order accuracy, L-stability, and asymptotic preservation. In the linear constant-coefficient case, the face-based derivative closure satisfies $\widetilde{\mathcal G}_h=\mathcal G_h^{\mathrm{tr}}=\mathcal L_h^2$ exactly, although it is assembled without squaring the residual matrix. The temporal analysis gives fourth-order local consistency, and the stability function is shown to be A-stable and L-stable exactly for $C\in[C_-,C_+]$; at $C_q=5/183$ the leading $1/z$ coefficient vanishes, giving $R(z)=314/(5z

Load-bearing premise

The load-bearing premise is that for nonlinear fluxes the ADER face-based approximation $\widetilde{\mathcal G}_h$ stays within $O(h^p)$ of the exact discrete trajectory derivative $\mathcal G_h^{\mathrm{tr}}$; the paper proves this only for linear constant-coefficient discretizations, assumes it for nonlinear ones, and its nonlinear numerical test uses the exact derivative rather than the ADER closure.

Editorial extensions

If this is right

  • Fourth order and L-stability can be obtained with two stages and two $N$-unknown sequential implicit solves, a combination unavailable to classical two-stage Runge–Kutta pairs.
  • The special parameter $C_q=5/183$ gives quadratic stiff decay, $R(z)=314/(5z^2)+O(z^{-3})$ as $z\to-\infty$, and in the two-dimensional divergence-free fast-mode experiment five steps at $\Delta t/\delta=200$ drive the mode to machine zero.
  • For fixed compatible divergence–gradient spaces, the full-step map satisfies $\|R(\Delta t A_{h,\delta})-\mathcal E_h R(\Delta t \mathcal L_{D,h})\mathcal P_h\|=O(\delta)$, so the method is asymptotic-preserving without requiring $\Delta t=O(\delta)$.
  • Uniform accuracy in the relaxation parameter is preparation-dependent: exact slow-manifold data and third-order slow-manifold-prepared data are uniformly fourth order, while limiting-equilibrium data have a first-order full-state barrier and unprepared data have no uniform full-state order; all bounded data recover uniform fourth order on any interval $[t_0,T]$ with $t_0>0$.
  • The stiff nonautonomous defect analysis predicts an intrinsic effective third-order window whenever $|\lambda|\Delta t\gg1$, so users in that regime should expect third-order global convergence until the time step is refined.

Reading between the lines

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

  • The nonlinear fourth-order claim is conditional on the trajectory-closure bound (71); a direct nonlinear test of the ADER face operator, rather than the exact trajectory derivative, would certify the bound or narrow the theorem to exact-closure implementations.
  • The sequential stage structure invites a shared preconditioner: since both stage systems contain the same diffusion-like macroscopic block, one preconditioner can serve both $N$-unknown solves, with stiffness-aware stopping for the nonlinear iterations.
  • The preparation-dependent accuracy results imply that initial data generation is part of the discretization in stiff relaxation: initializing with the equilibrium relation instead of the slow manifold costs a first-order full-state error in the initial layer no matter how high the temporal order of the integrator.
  • Since the asymptotic-preserving proof relies only on compatibility $\mathcal D_h=-\mathcal G_h^*$, L-stability, and finite-dimensional fixed spaces, the same full-step argument is likely to transfer to other symmetric hyperbolic relaxation systems with the same block structure, subject to constants that depend on the discrete spectrum.
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

3 major / 4 minor

Summary. The paper presents a fully implicit sequential two-stage fourth-order two-derivative time integrator for stiff transport–relaxation systems, closing a previously introduced temporal formula with a conservative finite-volume residual L_h and a discrete trajectory derivative. For the exact trajectory derivative G_h^tr = D L_h L_h, the authors prove fourth-order local consistency, A/L-stability, the asymptotic parameter C_q = 5/183 giving O(|z|^{-2}) stiff decay, and a Prothero–Robinson analysis showing an intrinsic third-order nonautonomous stiff window. They also prove, for fixed compatible linear finite-dimensional discretizations, a full-step asymptotic-preserving operator limit with O(δ) error and a preparation-dependent uniform-accuracy classification. Numerical experiments cover linear finite-volume ADER closure, nonlinear Jin–Xin convergence with exact trajectory derivatives, one- and two-dimensional order tests, diffusion limits, and modal initial-layer behavior. The main unresolved point is the nonlinear ADER face-based closure: the consistency bound (71) is assumed rather than proved, and the nonlinear experiments intentionally use the exact discrete trajectory derivative rather than the ADER provider.

Significance. If the claimed properties hold in the stated generality, the method is significant: it achieves fourth order and L-stability with only two sequential N-unknown implicit stages, gives an explicit asymptotic-preserving full-step limit at fixed h and Δt, and identifies a parameter regime with enhanced O(|z|^{-2}) stiff damping. Strengths include a self-contained order and stability analysis, an exact linear face-derivative closure identity verified to roundoff, detailed AP and uniform-accuracy theorems for the fixed-grid linear model, and numerical tests that match the predicted orders and rates. The principal weakness is that the nonlinear ADER trajectory-derivative closure, which the title and abstract foreground, is not established beyond an assumption; the nonlinear face-based ADER implementation is neither proved nor numerically exercised. In addition, a key algebraic identity in the A-stability proof appears to be misprinted. These issues are substantial but local in nature: the central derivation is coherent under the stated assumptions, and the gaps can be addressed within the manuscript's scope by supplying the missing nonlinear closure analysis/test and correcting the stabil

major comments (3)
  1. [§4.4, Eq. (71); §8.9, Eq. (139)] The nonlinear ADER closure bound (71), ||tilde G_h(Pi_h U) - G_h^tr(Pi_h U)|| <= C h^p, is assumed but not proved; Proposition 3 establishes exact equality only for linear constant-coefficient discretizations. The nonlinear convergence test in §8.9 uses the exact discrete trajectory derivative (139), not the face-based ADER provider, as the text after (139) explicitly states. Thus fourth-order behavior of the ADER-implemented nonlinear method is unverified. Since the paper's title and abstract present the ADER/CK predictor as the ingredient that closes the integrator for stiff transport–relaxation systems, this gap is load-bearing. Please either prove (71) under explicit nonlinearity/reconstruction assumptions, add a nonlinear test using the face-based tilde G_h, or clearly state the nonlinear ADER claim as conditional.
  2. [§4.4, Theorem 4, Eq. (72)] The fixed-h temporal fourth-order statement requires eta_G = O(Delta t^3), but Eq. (71) supplies only eta_G = O(h^p). At fixed h this is O(1), not O(Delta t^3), so Theorem 4 yields a combined O(h^p + Delta t^4) space–time error as h, Delta t -> 0, not fixed-h temporal fourth order. The sentence 'the ADER closure condition (71) gives eta_G = O(h^p), which is absorbed into the combined O(h^p + Delta t^4) space–time error' should be made precise: the absorbed term is Delta t * O(h^p) and is bounded by T h^p only under a bounded-Delta t convention. As written, it risks overstating the consequence of (71).
  3. [§3.3, Eq. (31) and Eq. (33)] There appears to be a sign/parenthesis error in the stability-function formulas. Read literally, Eq. (31) gives R(0) = -1, contradicting Eq. (30), which gives R(0)=1. If the intended numerator is the negative of the displayed bracket, then Eq. (33) still cannot be correct: at y=0 one must have |D_C(0)|^2 - |N_C(0)|^2 = 0, whereas Eq. (33) gives 9360C^2 - 600C + 8 (or its negative, depending on the parenthesization). Since Theorem 2's A-stability proof depends directly on Eq. (33), this is a load-bearing algebraic step. Please correct the formulas and supply a derivation of the imaginary-axis identity.
minor comments (4)
  1. [§3.3, Eq. (31)] Please add parentheses to the numerator of the closed-form stability function so that R(0)=1 is consistent with Eq. (30).
  2. [§6.1, Lemma 6] The proof of (108) invokes the 'strong maximum principle' for the ray x >= x_0. Since R(-x) is a real-analytic function of a real variable, the maximum principle is not directly applicable; continuity, R(-x) -> 0, and |R(-x)| <= 1 with R not identically 1 on the ray give the needed sup < 1. Please rephrase.
  3. [§4.3 and Algorithm 1] The notation G_h is used both for the exact discrete trajectory derivative G_h^tr and for the ADER approximation tilde G_h. Remark 1 helps, but the subsequent formulas (60), (134)–(135), and Algorithm 1 should consistently distinguish the two operators to avoid ambiguity.
  4. [§8.9] The sentence 'The reference uses 128 steps' should specify whether the reference is the exact semi-discrete solution or a fine-step numerical solution, and how the errors in Table 9 are computed.

Circularity Check

0 steps flagged · score 1.0 of 10

No material circularity: core order/L-stability/AP claims are derived from first principles; the unproven nonlinear ADER bound is an admitted correctness gap, not a circular reduction.

full rationale

I walked the claimed derivation chain. The temporal method (stages (16)-(17), coefficients (18)) and its order analysis are re-derived in Theorem 1; the stability function R(z) is computed from the scheme and A/L-stability is re-proven in Theorem 2 without importing the result from ref. [6]. The special parameter C_q=5/183 is obtained in Proposition 1 by setting the coefficient of 1/z in the rational expansion to zero (Eqs. (36)-(38)), not by fitting data to the claimed prediction. The AP limit in Theorem 5 follows from the block decomposition of the linear semi-discrete operator and the L-stability estimate; it does not assume the target operator limit. The linear trajectory-derivative identity (Proposition 3, Eq. (68)) is proven for the linear finite-volume assembly and is independently audited numerically (||tilde G_h - L_h^2|| ~ 1e-11); it is not a definitional equivalence. The only places where support is missing are explicitly flagged by the paper itself: Eq. (71) is an assumed trajectory-closure bound ||tilde G_h - G_tr_h|| <= C h^p for nonlinear discretizations, not a proved consequence; Section 11 says 'A general nonlinear face-based ADER closure still requires a verified bound relative to DL_h L_h.' The nonlinear order test in Section 8.9 uses the exact discrete trajectory derivative (Eq. (139)) and the text says 'It is not presented as a nonlinear face-based ADER closure test.' These are limitations in the nonlinear ADER claim, and Theorem 4's fixed-h temporal statement with eta_G=O(Delta t^3) is not supplied by (71)'s O(h^p); but a missing or assumed estimate is a correctness/verification gap, not circularity, because the target result is not used to define the assumption. I therefore find no circular step; the paper's own scope statements count against an inflated reading of its claims but do not make the derivation self-referential.

Assumptions & free parameters 1 free parameters · 5 assumptions · 0 invented entities

No new physical entities are postulated. The trajectory derivative G_tr_h = DL_h L_h is a standard discrete object; the ADER predictor is an existing technique. The paper's main 'invention' is the pairing (L_h, tilde G_h) and the parameter C_q, both mathematical constructs without independent physical falsifiability.

free parameters (1)
  • C = interval [C-, C+] with C-=(25-sqrt(105))/780, C+=(25+sqrt(105))/780; special values C-, Cq=5/183, C+
    Free parameter of the two-derivative method controlling order constant and stiff decay. The analysis proves L-stability on the interval; Cq is derived to cancel the 1/z term, not fitted to data.
assumptions (5)
  • domain assumption Compatible discrete divergence and gradient satisfy D_h = -G_h^* (adjoint condition)
    Used in Lemma 1 for energy dissipation and throughout Section 5 for the AP slow-fast decomposition. Requires quadrature-compatible finite-volume operators; the nonperiodic experiment uses a staggered pair satisfying it exactly.
  • ad hoc to paper The ADER/CK trajectory-derivative provider satisfies the closure estimate (71): ||tilde G_h(Pi_h U) - G_tr_h(Pi_h U)|| <= C h^p
    Assumed for nonlinear discretizations; the linear constant-coefficient case is proven exactly (Prop 3), but nonlinear ADER closure is not proven or numerically tested. The fourth-order temporal theorem requires G_tr_h, so (71) is load-bearing.
  • domain assumption Residual and exact-trajectory consistency of the spatial discretization, (69)-(70)
    Standard assumption for a consistent finite-volume scheme; used in Theorem 4 for space-time error.
  • domain assumption Exact solution of the implicit stage systems in the AP theorem
    Theorem 5 assumes exact stage solves; Proposition 4 extends to inexact solves with an implementation tolerance floor.
  • standard math Smoothness of L (L in C^4) for the order theorem
    Required for the Taylor expansion and implicit-function theorem in Theorem 1.

how reviews work

0 comments
Cite this review

Pith. "Pith review of An L-Stable Sequential Two-Stage Fourth-Order Method with ADER Trajectory Derivatives for Stiff Transport--Relaxation Systems." pith.science (2026). https://pith.science/paper/NF47CGAE

@misc{pith2026260803256,
  author       = {Pith},
  title        = {Pith review of: An L-Stable Sequential Two-Stage Fourth-Order Method with ADER Trajectory Derivatives for Stiff Transport--Relaxation Systems},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NF47CGAE}},
  note         = {Machine review of arXiv:2608.03256}
}
abstract

A fully implicit two-stage fourth-order two-derivative time discretization was introduced previously as a temporal method. This paper closes that sequential integrator for stiff transport--relaxation equations by pairing a conservative finite-volume residual $\mathcal L_h$ with its discrete trajectory derivative $\mathcal G_h^{\rm tr}=D\mathcal L_h\,\mathcal L_h$. An ADER/Cauchy--Kowalevski predictor provides interface states and physical time derivatives; differentiating the same numerical flux and taking shared face differences yields a conservative approximation $\widetilde{\mathcal G}_h$. For linear constant-coefficient balance laws, $\widetilde{\mathcal G}_h=\mathcal G_h^{\rm tr}=\mathcal L_h^2$ exactly, although the derivative operator is assembled independently rather than by squaring the residual matrix. For nonlinear discretizations, the fourth-order temporal theory applies to $\mathcal G_h^{\rm tr}$, while a trajectory-closure consistency estimate controls the ADER approximation. The two unknown stage vectors are solved successively through two $N$-unknown systems. The completed step is fourth order and L-stable; the parameter $C_q=5/183$ cancels the leading inverse-power term and changes the deep-stiff amplification from $O(|z|^{-1})$ to $O(|z|^{-2})$. For fixed compatible spatial spaces, a slow--fast decomposition proves a full-step asymptotic-preserving operator limit with an $O(\delta)$ estimate and gives a preparation-dependent uniform-accuracy classification. Linear finite-volume, nonlinear relaxation, one- and two-dimensional damping, diffusion-limit, and modal experiments verify the corresponding closure, accuracy, stability, and singular-limit claims within their stated scopes.

Figures

Figures reproduced from arXiv: 2608.03256 by the authors.

Figure 1
Figure 1. Absolute-stability contours for the accuracy endpoint C−, the quadratic-decay value Cq = 5/183, and the damping endpoint C+ [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Small-error and enhanced-damping parameter choices inside the same L-stable family. The Cq curve eventually follows O(|z|−2 ). among the three representative values. When the solution contains a strongly damped initial layer or transverse relaxation mode, Cq can be preferable because the asymptotic amplification is two powers smaller. The endpoint C+ remains useful as a conservative member of the proven interval and… view at source ↗
Figure 3
Figure 3. One-dimensional fourth-order temporal convergence [PITH_FULL_IMAGE:figures/full_fig_p020_3.png] view at source ↗
Figures from the paper (17 more)
Figure 4
Figure 4. Figure 4: Fourth-order convergence when the stage equations use the ADER/CK face flux-time-derivative operator Geh rather than a preformed square of the residual matrix. 8.2 Finite-volume ADER trajectory-derivative coupling The preceding test isolates the time formula with a mat…
Figure 5
Figure 5. Figure 5: Stiff decay on the negative real axis [PITH_FULL_IMAGE:figures/full_fig_p021_5.png]
Figure 6
Figure 6. Figure 6: High-precision Prothero–Robinson test. Extreme nonautonomous stiffness exhibits the analytically predicted third-order window, while fixed moderate stiffness recovers fourth order. 8.3 Scalar stiff decay and the enhanced parameter [PITH_FULL_IMAGE:figures/full_fig_p02…
Figure 7
Figure 7. Figure 7: Maximum-in-time uniform-accuracy scan over 10−8 ≤ δ ≤ 10−2 . The third-order Chapman–Enskog preparation is a practical approximation to the exact discrete slow subspace and asymptotically restores full-state fourth order [PITH_FULL_IMAGE:figures/full_fig_p022_7.png]
Figure 8
Figure 8. Figure 8: Uniform fourth-order recovery after a fixed positive time for limiting-equilibrium and unprepared data. 8.5 Uniform accuracy, constructive preparation, and post-layer recovery The mode-level experiment corresponding to Section 6 uses a = 0.7, wave number σ = 4, tf = 0.…
Figure 9
Figure 9. Figure 9: Preparation-dependent uniform accuracy for a smooth three-mode PDE state [PITH_FULL_IMAGE:figures/full_fig_p023_9.png]
Figure 10
Figure 10. Figure 10: Cattaneo pulse for δ = 10−1 , 10−2 , 10−4 . Solid curves are the proposed method and dashed curves are exact semi￾discrete references. Multi-mode PDE verification. To check that the preparation-dependent orders are not an artifact of a single Fourier block, we repeat …
Figure 11
Figure 11. Figure 11: Nonperiodic Cattaneo eigenmode with homogeneous boundary fluxes and a compatible staggered divergence–gradient pair. 8.8 Classical Goldstein–Taylor diffusion limit A discontinuous benchmark is computed on [−1, 1] with N = 160, a = 1/2, tf = 0.04, and ∆t = 2 × 10−4 . T…
Figure 12
Figure 12. Figure 12: Goldstein–Taylor/Cattaneo diffusion-limit benchmark. 24 [PITH_FULL_IMAGE:figures/full_fig_p024_12.png]
Figure 13
Figure 13. Figure 13: Fourth-order convergence for the nonlinear diffusive Jin–Xin problem. The reference uses 128 steps. Both implicit stages use the exact Jacobians associated with (139) [PITH_FULL_IMAGE:figures/full_fig_p025_13.png]
Figure 14
Figure 14. Figure 14: Nonlinear diffusion-limit evidence: the Jin–Xin scalar approaches the same-grid viscous Burgers solution at approximately O(δ). 9 Two-dimensional numerical experiments 9.1 Two-dimensional temporal convergence A periodic 22 × 22 mode is evolved with a = 0.5, δ = 10−2 ,…
Figure 15
Figure 15. Figure 15: Two-dimensional fourth-order temporal convergence. 26 [PITH_FULL_IMAGE:figures/full_fig_p026_15.png]
Figure 16
Figure 16. Figure 16: Two-dimensional stiff Cattaneo Gaussian field [PITH_FULL_IMAGE:figures/full_fig_p027_16.png]
Figure 17
Figure 17. Figure 17: Time history of the divergence-free fast flux mode. 9.3 Divergence-free fast-mode initial layer Let ψ = sin x sin y and initialize qx = ∂yψ, qy = −∂xψ, (143) with zero scalar perturbation. This flux is divergence-free and therefore satisfies the exact decoupled equati…
Figure 18
Figure 18. Figure 18: Initial and final flux magnitudes on a common color scale. Gauss4 and Crank–Nicolson retain the spatial pattern; the proposed method removes it. 9.4 Direct PDE effect of the quadratic-decay parameter The previous comparison used the upper endpoint C+. To isolate the p…
Figure 19
Figure 19. Figure 19: Fast-mode amplitude for the three admissible parameters in the same two-dimensional relaxation initial layer. 10 AP verification and synthesis of the numerical evidence 10.1 Direct verification of the AP operator theorem The one-step matrix R(∆tAh,δ) is compared direc…
Figure 20
Figure 20. Figure 20: O(δ) convergence of the full relaxation step to the equilibrium embedding of the discrete diffusion step. assembled. For nonlinear discretizations, the temporal theorem applies to G tr h and the ADER replacement is controlled by an explicit trajectory-closure consiste…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

33 extracted references · 32 canonical work pages

  1. [6]

    An L-stable implicit two-stage fourth-order temporal discretization scheme for Lax–Wendroff-type solvers applied to stiff problems,

    Z. Huo, “An L-stable implicit two-stage fourth-order temporal discretization scheme for Lax–Wendroff-type solvers applied to stiff problems,” arXiv:2512.01628v4, 2026

    arXiv 2026

  2. [1]

    Second derivative methods with Runge–Kutta stability,

    J. C. Butcher and Z. Jackiewicz, “Second derivative methods with Runge–Kutta stability,” Numer. Algorithms , 40 (2005), 415–429

    work page 2005

  3. [2]

    On explicit two-derivative Runge–Kutta methods,

    R. P. K. Chan and A. Y. J. Tsai, “On explicit two-derivative Runge–Kutta methods,” Numer. Algorithms , 53 (2010), 171–194

    work page 2010

  4. [3]

    High-order multiderivative time integrators for hyperbolic conservation laws,

    D. C. Seal, Y. Güçlü, and A. J. Christlieb, “High-order multiderivative time integrators for hyperbolic conservation laws,” J. Sci. Comput. , 60 (2014), 101–140

    work page 2014

  5. [4]

    Jacobian-free implicit multiderivative Runge–Kutta methods for stiff systems of ODEs,

    J. Chouchoulis and J. Schütz, “Jacobian-free implicit multiderivative Runge–Kutta methods for stiff systems of ODEs,” Appl. Numer. Math. , 196 (2024), 45–61

    work page 2024

  6. [5]

    High order strong stability preserving multiderivative implicit and IMEX Runge– Kutta methods with asymptotic preserving properties,

    S. Gottlieb, Z. J. Grant, J. Hu, and R. Shu, “High order strong stability preserving multiderivative implicit and IMEX Runge– Kutta methods with asymptotic preserving properties,” SIAM J. Numer. Anal. , 60 (2022), 423–449. 29

    work page 2022

  7. [7]

    A Compact Two-Stage Fourth-Order Two-Derivative IMEX Method with Mixed Compatibility, High Implicit-Solve Efficiency,and Enhanced Stiff Decay

    Z. Huo, “A compact two-stage fourth-order two-derivative IMEX method with mixed compatibility, high implicit-solve efficiency, and enhanced stiff decay,” arXiv:2606.23319v1, 2026

    work page Pith review arXiv 2026

  8. [8]

    Arbitrary-Order Pad\'e-Closed Anchored Two-Derivative Time Discretizations: $s$ Active Stages, Order $2s$, and $L$-Stability

    Z. Huo, “Arbitrary-order Padé-closed anchored two-derivative time discretizations: s active stages, order 2s, and L-stability,” arXiv:2607.24592v1, 2026

    work page Pith review arXiv 2026

Show all 33 references
  1. [9]

    A two-stage fourth order time-accurate discretization for Lax–Wendroff type flow solvers I: Hyperbolic conservation laws,

    J. Li and Z. Du, “A two-stage fourth order time-accurate discretization for Lax–Wendroff type flow solvers I: Hyperbolic conservation laws,” SIAM J. Sci. Comput. , 38 (2016), A3046–A3069

    2016

  2. [10]

    A Hermite WENO reconstruction for fourth order temporal accurate schemes based on the GRP solver for hyperbolic conservation laws,

    Z. Du and J. Li, “A Hermite WENO reconstruction for fourth order temporal accurate schemes based on the GRP solver for hyperbolic conservation laws,” J. Comput. Phys. , 355 (2018), 385–396

    2018

  3. [11]

    A two-stage fourth order time-accurate discretization for Lax–Wendroff type flow solvers II: High order numerical boundary conditions,

    Z. Du and J. Li, “A two-stage fourth order time-accurate discretization for Lax–Wendroff type flow solvers II: High order numerical boundary conditions,” J. Comput. Phys. , 369 (2018), 125–147

    2018

  4. [12]

    Uniformly accurate schemes for hyperbolic systems with relaxation,

    R. E. Caflisch, S. Jin, and G. Russo, “Uniformly accurate schemes for hyperbolic systems with relaxation,” SIAM J. Numer. Anal., 34 (1997), 246–281

    1997

  5. [13]

    On a class of uniformly accurate IMEX Runge–Kutta schemes and applications to hyperbolic systems with relaxation,

    S. Boscarino and G. Russo, “On a class of uniformly accurate IMEX Runge–Kutta schemes and applications to hyperbolic systems with relaxation,” SIAM J. Sci. Comput. , 31 (2009), 1926–1945

    2009

  6. [14]

    On the uniform accuracy of implicit–explicit backward differentiation formulas for stiff hyperbolic relaxation systems and kinetic equations,

    J. Hu and R. Shu, “On the uniform accuracy of implicit–explicit backward differentiation formulas for stiff hyperbolic relaxation systems and kinetic equations,” Math. Comp. , 90 (2021), 641–670

    2021

  7. [15]

    Uniform accuracy of implicit–explicit Runge–Kutta methods for linear hyperbolic relaxation systems,

    Z. Ma and J. Huang, “Uniform accuracy of implicit–explicit Runge–Kutta methods for linear hyperbolic relaxation systems,” J. Sci. Comput. , 104 (2025), Article 60

    2025

  8. [16]

    The relaxation schemes for systems of conservation laws in arbitrary space dimensions,

    S. Jin and Z. Xin, “The relaxation schemes for systems of conservation laws in arbitrary space dimensions,” Commun. Pure Appl. Math. , 48 (1995), 235–276

    1995

  9. [17]

    Numerical schemes for hyperbolic conservation laws with stiff relaxation terms,

    S. Jin and C. D. Levermore, “Numerical schemes for hyperbolic conservation laws with stiff relaxation terms,” J. Comput. Phys. , 126 (1996), 449–467

    1996

  10. [18]

    Diffusive relaxation schemes for multiscale discrete-velocity kinetic equations,

    S. Jin, L. Pareschi, and G. Toscani, “Diffusive relaxation schemes for multiscale discrete-velocity kinetic equations,” SIAM J. Numer. Anal. , 35 (1998), 2405–2439

    1998

  11. [19]

    Uniformly accurate diffusive relaxation schemes for multiscale transport equations,

    S. Jin, L. Pareschi, and G. Toscani, “Uniformly accurate diffusive relaxation schemes for multiscale transport equations,” SIAM J. Numer. Anal. , 38 (2000), 913–936

    2000

  12. [20]

    An asymptotic-preserving well-balanced scheme for the hyperbolic heat equations,

    L. Gosse and G. Toscani, “An asymptotic-preserving well-balanced scheme for the hyperbolic heat equations,” C. R. Math. Acad. Sci. Paris , 334 (2002), 337–342

    2002

  13. [21]

    A form of heat conduction equation which eliminates the paradox of instantaneous propagation,

    C. Cattaneo, “A form of heat conduction equation which eliminates the paradox of instantaneous propagation,” Comptes Rendus, 247 (1958), 431–433

    1958

  14. [22]

    Les paradoxes de la théorie continue de l’équation de la chaleur,

    P. Vernotte, “Les paradoxes de la théorie continue de l’équation de la chaleur,” Comptes Rendus , 246 (1958), 3154–3155

    1958

  15. [23]

    ADER: arbitrary high order Godunov approach,

    V. A. Titarev and E. F. Toro, “ADER: arbitrary high order Godunov approach,” J. Sci. Comput. , 17 (2002), 609–618

    2002

  16. [24]

    Finite volume schemes of very high order of accuracy for stiff hyperbolic balance laws,

    M. Dumbser, C. Enaux, and E. F. Toro, “Finite volume schemes of very high order of accuracy for stiff hyperbolic balance laws,” J. Comput. Phys. , 227 (2008), 3971–4001

    2008

  17. [25]

    A unified framework for the construction of one-step finite volume and discontinuous Galerkin schemes on unstructured meshes,

    M. Dumbser, D. S. Balsara, E. F. Toro, and C.-D. Munz, “A unified framework for the construction of one-step finite volume and discontinuous Galerkin schemes on unstructured meshes,” J. Comput. Phys. , 227 (2008), 8209–8253

    2008

  18. [26]

    A direct Eulerian GRP scheme for compressible fluid flows,

    M. Ben-Artzi, J. Li, and G. Warnecke, “A direct Eulerian GRP scheme for compressible fluid flows,” J. Comput. Phys. , 218 (2006), 19–43

    2006

  19. [27]

    Hairer and G

    E. Hairer and G. Wanner, Solving Ordinary Differential Equations II: Stiff and Differential-Algebraic Problems , 2nd ed., Springer, 1996

    1996

  20. [28]

    On the stability and accuracy of one-step methods for solving stiff systems of ordinary differential equations,

    A. Prothero and A. Robinson, “On the stability and accuracy of one-step methods for solving stiff systems of ordinary differential equations,” Math. Comp. , 28 (1974), 145–162

    1974

  21. [29]

    The Prothero and Robinson example: Convergence studies for Runge–Kutta and Rosenbrock–Wanner methods,

    J. Rang, “The Prothero and Robinson example: Convergence studies for Runge–Kutta and Rosenbrock–Wanner methods,” Appl. Numer. Math. , 108 (2016), 37–56

    2016

  22. [30]

    Jacobian-free Newton–Krylov methods: a survey of approaches and applications,

    D. A. Knoll and D. E. Keyes, “Jacobian-free Newton–Krylov methods: a survey of approaches and applications,” J. Comput. Phys., 193 (2004), 357–397

    2004

  23. [31]

    GMRES: a generalized minimal residual algorithm for solving nonsymmetric linear systems,

    Y. Saad and M. H. Schultz, “GMRES: a generalized minimal residual algorithm for solving nonsymmetric linear systems,” SIAM J. Sci. Stat. Comput. , 7 (1986), 856–869

    1986

  24. [32]

    A flexible inner-outer preconditioned GMRES algorithm,

    Y. Saad, “A flexible inner-outer preconditioned GMRES algorithm,” SIAM J. Sci. Comput. , 14 (1993), 461–469

    1993

  25. [33]

    E. F. Toro, Riemann Solvers and Numerical Methods for Fluid Dynamics , 3rd ed., Springer, 2009. 30

    2009

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.