Conditional a priori error estimates of finite volume and Runge-Kutta discontinuous Galerkin methods with abstract limiting for hyperbolic systems of conservation laws in 1D

Fabio Leotta

REVIEW 3 major objections 5 minor 1 cited by

Conditional a priori error estimates of finite volume and Runge-Kutta discontinuous Galerkin methods with abstract limiting for hyperbolic systems of conservation laws in 1D

T0 review · 3 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 Monotone finite volume and two-stage Runge-Kutta discontinuous Galerkin schemes for one-dimensional hyperbolic systems satisfy the conditions for a conditional a priori L1 error bound of order $h^{1/3}$.

desk verdict Solid, honest conditional result: verifies every Bressan-Chiri-Shen hypothesis except the BV-control assumption, which remains open for all concrete schemes. read the letter →

arxiv 2506.13221 v2 pith:NQ5VVKIB submitted 2025-06-16 math.NA cs.NA

Fabio Leotta This is my paper · ORCID

classification math.NAcs.NA MSC 65M0865M1265M6035L65

keywords hyperbolicsystemsofconservationlawsfinitevolumemethodsRunge-KuttadiscontinuousGalerkinentropystabilityapriorierrorestimatesBVslopelimitingweakconsistency

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 proves that a broad class of numerical methods for one-dimensional systems of hyperbolic conservation laws—finite volume schemes with strictly monotone numerical fluxes and two-stage Runge-Kutta discontinuous Galerkin (RKDG) schemes with abstract limiting—satisfy the three structural conditions (Lipschitz continuity in time, weak consistency, weak entropy stability) required by the error-estimation framework of [4]. Under the CFL condition (4.23), any such scheme inherits the framework's a priori error estimate: if the numerical solution is uniformly small in BV and has oscillation strength $h$ outside narrow $h^{2/3}$-tubes around the exact shock curves (the BV-control hypothesis), the $L^1$ error at time $T$ is bounded by a constant times $h^{1/3}$. The result is conditional because the BV-control hypothesis is assumed, not proved; the paper calls it a working hypothesis and notes a known Godunov instability [5] where it fails. If the hypothesis holds for a given scheme, this yields a quantitative convergence-rate result for general systems with shocks, in the spirit of a quantitative Lax-Wendroff theorem.

What carries the argument

The load-bearing objects are the entropy variables $v = \nabla\eta(u)$ and the numerical flux expressed as a function of them. The scheme's flux is required to be uniformly strictly monotone in these variables (left Jacobian symmetric positive definite, right Jacobian symmetric negative definite), which yields positive definite quadratic dissipative terms in the entropy balance. The key identity is the second-order consistency relation (3.4), which makes the two numerical entropy fluxes $Q^r$ and $Q^l$ coincide (equation (4.17)) and reduces the entropy balance to a competition between dissipation from flux monotonicity and anti-dissipation from forward Euler time stepping. The CFL condition (4.23) states the precise balance: it requires $\tau/h$ to be bounded by a constant proportional to the square of the ratio of minimal to maximal eigenvalue of the entropy Hessian, times a factor involving the minimal and maximal eigenvalues of the flux Jacobians. For the DG part, the abstract limiter must conserve means and satisfy (6.4), the slope-boundedness condition, so that limiting perturbations and projection errors are absorbed into $h$-scaled terms.

What would settle it

Compute the left-hand sides of (2.4)-(2.5) and the $L^1$ error at a fixed time for a monotone finite volume scheme with HLL entropy fix on a sequence of meshes with the CFL condition (4.23), for a piecewise Lipschitz exact solution with a single shock. If the error decays slower than $h^{1/3}$ while the BV-control quantities stay bounded, or if the BV-control quantities themselves grow, the conditional claim has no instance in this test case.

Watch

Extended reading notes

Core claim

The central claim is Corollaries 4.3 and 6.3: for a strictly hyperbolic system with a strictly convex entropy, a finite volume solution computed with a first- and second-order consistent numerical flux that is uniformly strictly monotone in the entropy variables, or a two-stage RKDG solution with a slope limiter that conserves cell means and does not increase $L^\infty$ slopes, satisfies the error estimate $\|u(T,\cdot) - S_T(u_0)\|_{L^1} \lesssim h^{1/3}$, provided the BV-control hypothesis (equations (2.4)-(2.5)) holds and the time step obeys $\tau \leq c h$ with $c$ from (4.23). The proof verifies the three hypotheses of the abstract framework—time-Lipschitz continuity, weak consistency, and weak entropy stability—by rewriting the numerical flux in entropy variables and showing that its strict monotonicity produces dissipative terms that dominate the anti-dissipative forward-Euler terms under the CFL condition; for the DG case, projection errors and limiter perturbations are controlled using bounded slopes and total variation. The author presents this as a conditional convergence result: the BV-control hypothesis, concerning uniform BV bounds and local oscillation strength $h$ outside $h^{2/3}$ shock-tracing tubes, is explicitly a working hypothesis.

Load-bearing premise

The numerical solution is assumed to be uniformly small in BV and to oscillate by at most $h$ across each cell and cell boundary outside the $h^{2/3}$-thin shock tubes (the BV-control hypothesis (2.4)-(2.5)); the paper does not prove this for any concrete scheme and cites a known instability where it fails.

Editorial extensions

If this is right

Any finite volume scheme with a numerical flux that is first- and second-order consistent and strictly monotone in entropy variables—such as HLL with entropy fix or Lax-Friedrichs/Rusanov with suitable stabilization—falls under Corollary 4.3.
For RKDG schemes, any slope limiter satisfying the abstract properties (mean conservation and no increase of $L^\infty$ slopes) is compatible with the estimate, so the result covers a parametrized family of limiters rather than one specific recipe.
The error rate $h^{1/3}$ comes with the explicit CFL restriction (4.23), which degrades when the entropy Hessian is ill-conditioned; near vacuum states of the isothermal Euler system, the constant becomes very small, forcing very small time steps.
If the BV-control hypothesis is verified for a particular scheme, the same theorem immediately yields a true a priori convergence rate for that scheme for general systems.

Reading between the lines

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

The BV-control hypothesis is likely the real bottleneck for unconditional rates: the $h^{2/3}$ tube width and $h$ oscillation strength are coupled to the $h^{1/3}$ rate, so a scheme that provably satisfies (2.4)-(2.5) would immediately be the first general-system scheme with a rigorously proven rate.
The framework suggests a numerical diagnostic: monitoring the quantities in (2.5) on meshes of decreasing size could tell practitioners whether a given flux/limiter combination lies in the regime where convergence at rate $h^{1/3}$ is guaranteed.
One could test the sharpness of the $h^{1/3}$ rate by constructing an example satisfying the hypotheses where oscillation strength saturates the bound $h$; if the actual error decays faster, the framework is not sharp, and if slower, some hidden assumption is violated.
For DG, the limiter conditions (6.3)-(6.4) are purely abstract; a natural next step is to check whether standard limiters (minmod-type, moment-based) satisfy them uniformly, which would make the estimate concrete.

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.

Desk Editor's Note

read the letter

Short version: this is a solid, honest paper that does real work verifying all the Bressan-Chiri-Shen hypotheses except the one that matters most—BV-control—and it does not pretend otherwise. The entropy-variable monotonicity argument (Section 4) is a genuine technical contribution, and the extension to two-stage RKDG with abstract limiting (Section 6) is nontrivial, even if more compressed. If someone later proves the BV-control bound for one of these schemes, Corollaries 4.3 and 6.3 become a real convergence-rate result for systems with shocks.

I largely agree with the reader's take. The FV analysis is coherent: the entropy balance (4.12)-(4.22) and the CFL condition (4.23) check out, and the anti-dissipative term is tracked carefully. The second-order consistency condition (3.4) is a nice way to get a consistent numerical entropy flux without relying on exact Riemann solvers. The paper's claims are honestly labeled conditional; Remark 2.1 explicitly calls the BV-control property a working hypothesis and cites the Bressan-Jenssen-Baiti Godunov instability.

The soft spots are real but proportionate. The BV-control hypothesis (2.4)-(2.5) is not proved for any concrete scheme, and the stress-test note is right that if it fails, the corollaries have no content. That is the load-bearing assumption, and the paper does not reduce it. The RKDG section is more compressed: the conversion in (6.12) via CFL is sketched, and Remark 6.1 concedes that the positivity of the quadratic terms after limiting is only proved in the scalar case. The limiter assumption (6.4) controls limited slopes relative to input slopes, but it does not by itself guarantee the local oscillation condition (2.5) across the two Runge-Kutta stages. These are gaps, but they are gaps in a conditional statement that is already explicitly conditional.

Net: this paper is for numerical analysts working on hyperbolic systems, especially those interested in the Bressan-Chiri-Shen framework or in proving BV stability. It deserves a serious referee. I would accept it for peer review; the main question for the referee is whether the conditional result is significant enough given the unproved hypothesis. I think it is, as a stepping stone—it isolates exactly what remains open. I'd probably cite it for the FV entropy-stability lemma (Lemma 4.2) and the flux conditions.

Referee Report

3 major / 5 minor

Summary. The paper derives conditional a priori error estimates in L∞L1 with rate h^{1/3} for a class of finite volume (FV) and Runge-Kutta discontinuous Galerkin (RKDG) methods for one-dimensional hyperbolic systems, following the framework of Bressan–Chiri–Shen. It proves that the FV and RKDG schemes satisfy the weak consistency (LC), (WC), and weak entropy stability (WES) properties under a CFL condition, assuming a working BV-control hypothesis (2.4)-(2.5). The main results are Corollary 4.3 for FV and Corollary 6.3 for RKDG.

Significance. The paper contributes a rigorous verification of the structural hypotheses required by the Bressan–Chiri–Shen error-estimation framework for a broad class of monotone fluxes and for DG methods with abstract limiters. The FV entropy analysis, in particular, carefully balances anti-dissipative time-discretization effects against flux dissipation. The conditional nature is honestly disclosed, and the paper explicitly cites the known Godunov instability as a limitation. However, the applicability of the main theorem is contingent on the unresolved BV-control hypothesis, and the RKDG proof in Section 6 is too compressed to be fully checked.

major comments (3)

[Section 6, Lemma 6.2] The verification of (LC), (WC), and (WES) for the two-stage RKDG scheme is incomplete. The argument after (6.6) only estimates the limiter perturbation terms I^1 and I^2 and then refers to the forward Euler analysis of Section 5 without assembling the full entropy inequality (2.3). In particular, the transition from (6.11) to (6.12) uses a "generic CFL condition" whose precise relation to (4.23) is not stated, and the treatment of the two RK stages as two forward Euler steps is asserted rather than demonstrated. Corollary 6.3 therefore does not follow from the written proof.
[Section 6, equations (6.9)-(6.11)] The limiter assumption (6.4) bounds only the L∞ norm of the limited slope, while the bound on I^2 in (6.11) requires an L2 bound on e u - u_j relative to u - u_j. This L2 bound is not a consequence of (6.3)-(6.4) for an abstract limiter; the proof needs an additional assumption or a proof that the limiter does not increase the L2 deviation from the cell mean.
[Section 2 and Corollary 6.3] The error estimate (2.6) is entirely conditional on the BV-control hypothesis (2.4)-(2.5), which is not verified for any of the schemes considered. For the RKDG case, the preservation of the local oscillation condition (2.5) under the abstract limiter across the two stages is not analyzed. Since the Bressan–Jenssen–Baiti counterexample shows that (2.4) can fail for a Godunov scheme, the set of schemes to which the results apply is currently empty; the paper should at least discuss what additional evidence would be needed to satisfy (BV) for the schemes considered.

minor comments (5)

[Section 2, Remark 2.1] The statement "cf. the multitude of examples presented in [17]" is vague; it would be helpful to cite specific schemes or sections of Toro's book that exhibit BV stability numerically.
[Section 3, Example 3.2] The borderline case s^- = 0 or s^+ = 0 in the HLL flux is discussed, but the subsequent statement about entropy fix (Example 3.3) could clarify how the threshold δ handles these cases quantitatively.
[Section 4.3, equation (4.19)] The notation λ_max(|Df|) is used without defining the absolute value of a matrix; it should be replaced by the spectral radius or a norm.
[Section 6] The limiting operator is not formally defined; in particular, e u^{n+1} appears in (6.7) without having been introduced in the scheme (6.1)-(6.2). Define the limiter and its application to each stage.
[Throughout] There are several typos and awkward phrasings, e.g., "approximate 1 solutions" near the start of Section 2 and the incomplete sentence in Remark 6.1; a careful proofread is needed.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the h^{1/3} estimate is explicitly conditional on the unproved BV-control hypothesis, and the verified conditions (LC), (WC), (WES) are independent of the conclusion.

full rationale

The central results, Corollaries 4.3 and 6.3, are conditional on the BV-control hypothesis (2.4)-(2.5), which the paper explicitly labels a working hypothesis in Remark 2.1 and does not claim to prove for any concrete scheme. A theorem that states 'if (BV), then error ≲ h^{1/3}' is not circular merely because the hypothesis is unverified; the conclusion does not enter the hypothesis and the hypothesis is not a renamed version of the conclusion. The derivation chain relies on the external a posteriori framework of Bressan-Chiri-Shen [4], which is independent of the present paper, and Sections 4 and 6 verify (LC), (WC), and (WES) for finite volume and RKDG methods using consistency, monotonicity, entropy arguments, and explicit CFL conditions. No parameter is fitted to data, and no numerical output is relabeled as a prediction. The only self-citation, [13], is used for a standard discrete trace estimate with a Gauss-Radau interpolant in Section 5.1.2; this is a technical interpolation tool, not a load-bearing premise of the main theorem. If the BV-control hypothesis fails for a particular scheme, the corollaries simply do not apply; that is an unverified assumption and a correctness risk, not circular reasoning.

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

The paper introduces no fitted parameters and no new physical entities. It rests on the small-BV semigroup theory, strict monotonicity and second-order consistency of the numerical flux, the unproved BV-control hypothesis, and abstract limiter properties. The BV-control hypothesis is the dominant burden: the proofs of (LC), (WC), (WES) are independent of it, but the final convergence claim is not.

assumptions (5)

domain assumption Small-BV semigroup theory for (1.1) with Lipschitz dependence (1.4)-(1.6).
Invoked in Section 1 as the standard well-posedness background.
domain assumption Numerical flux is first- and second-order consistent and strictly monotone in entropy variables (conditions (i)-(iii), (3.4)).
These restrict the admissible FV/RKDG fluxes and enter directly into the entropy balance (4.12)-(4.22).
domain assumption BV-control hypothesis (2.4)-(2.5) holds for the numerical solution.
Explicitly called a working hypothesis in Remark 2.1; without it the error estimate in Theorem 2.2 does not apply.
domain assumption Exact solution is piecewise Lipschitz with finitely many isolated shocks and shock curves traceable to precision h^(2/3).
Needed for the abstract error estimate inherited from Bressan-Chiri-Shen (Theorem 2.2).
domain assumption The limiter preserves cell means (6.3) and bounds slopes in L∞ (6.4), with uniform slope boundedness assumed in Corollary 6.3.
Used in Section 6 to bound the limiter perturbations I1 and I2.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Conditional a priori error estimates of finite volume and Runge-Kutta discontinuous Galerkin methods with abstract limiting for hyperbolic systems of conservation laws in 1D." pith.science (2026). https://pith.science/paper/NQ5VVKIB

@misc{pith2026250613221,
  author       = {Pith},
  title        = {Pith review of: Conditional a priori error estimates of finite volume and Runge-Kutta discontinuous Galerkin methods with abstract limiting for hyperbolic systems of conservation laws in 1D},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NQ5VVKIB}},
  note         = {Machine review of arXiv:2506.13221}
}

abstract

We derive conditional a priori error estimates of a wide class of finite volume and Runge-Kutta discontinuous Galerkin methods with abstract limiting for hyperbolic systems of conservation laws in 1D via the verification of weak consistency and entropy stability, as recently proposed by Bressan et al.~\cite{BressanChiriShen21}. Convergence in $L^\infty L^1$ with rate $h^{1/3}$ is obtained under a time step restriction $\tau\leq ch$, provided the following conditions hold: the exact solution is piecewise Lipschitz continuous, its (finitely many and isolated) shock curves can be traced with precision $h^{2/3}$ and, outside of these shock tracing tubular neighborhoods the numerical solution -- assumed to be uniformly small in BV -- has oscillation strength $h$ across each mesh cell and cell boundary.

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

Convergence of entropy-conservative summation-by-parts discretizations to smooth solutions of hyperbolic conservation laws
math.NA 2026-07 accept novelty 6.0 of 10

Entropy-conservative diagonal-norm SBP flux-differencing schemes converge at order p to smooth solutions of general entropy-symmetrizable hyperbolic systems under periodic boundaries.

Reference graph

Works this paper leans on

17 extracted references · 17 canonical work pages · cited by 1 Pith paper

[1]

Vanishing viscosity solutions of nonlinear hyperbolic systems

Stefano Bianchini and Alberto Bressan. Vanishing viscosity solutions of nonlinear hyperbolic systems. Ann. of Math. (2) , 161(1):223–342, 2005

work page 2005
[2]

Bressan and H

A. Bressan and H. K. Jenssen. On the convergence of Godunov scheme for nonlinear hyperbolic systems.Chinese Ann. Math. Ser. B , 21(3):269–284, 2000

work page 2000
[3]

Hyperbolic systems of conservation laws , volume 20 of Oxford Lecture Series in Mathematics and its Applications

Alberto Bressan. Hyperbolic systems of conservation laws , volume 20 of Oxford Lecture Series in Mathematics and its Applications . Oxford University Press, Oxford, 2000. The one-dimensional Cauchy problem

work page 2000
[4]

A posteriori error estimates for numerical solutions to hyperbolic conservation laws

Alberto Bressan, Maria Teresa Chiri, and Wen Shen. A posteriori error estimates for numerical solutions to hyperbolic conservation laws. Arch. Ration. Mech. Anal. , 241(1):357–402, 2021

work page 2021
[5]

An instability of the Godunov scheme

Alberto Bressan, Helge Kristian Jenssen, and Paolo Baiti. An instability of the Godunov scheme. Comm. Pure Appl. Math. , 59(11):1604–1638, 2006

work page 2006
[6]

TVB Runge-Kutta local projection discontinuous Galerkin finite element method for conservation laws

Bernardo Cockburn, San Yih Lin, and Chi-Wang Shu. TVB Runge-Kutta local projection discontinuous Galerkin finite element method for conservation laws. III. One-dimensional systems.J. Comput. Phys., 84(1):90–113, 1989

work page 1989
[7]

Multiresolution schemes for conservation laws

Wolfgang Dahmen, Birgit Gottschlich-M¨ uller, and Siegfried M¨ uller. Multiresolution schemes for conservation laws. Numer. Math. , 88(3):399–443, 2001

work page 2001
[8]

Error control for a class of Runge-Kutta discontinuous Galerkin methods for nonlinear conservation laws

Andreas Dedner, Charalambos Makridakis, and Mario Ohlberger. Error control for a class of Runge-Kutta discontinuous Galerkin methods for nonlinear conservation laws. SIAM J. Numer. Anal. , 45(2):514–538, 2007

work page 2007

Show all 17 references

[9]

Convergence of the fractional step Lax-Friedrichs scheme and Godunov scheme for the isentropic system of gas dynamics

Xia Xi Ding, Gui Qiang Chen, and Pei Zhu Luo. Convergence of the fractional step Lax-Friedrichs scheme and Godunov scheme for the isentropic system of gas dynamics. Comm. Math. Phys. , 121(1):63–84, 1989

1989
[10]

Modeling, Simulation and Applications

Eduard Feireisl, M´ aria Luk´ acov´ a-Medvidov´ a, Hana Mizerov´ a, and Bangwei She.Numerical analysis of com- pressible fluid flows , volume 20 of MS&A. Modeling, Simulation and Applications . Springer, Cham, [2021] ©2021. ERROR ESTIMATES OF FV AND RKDG METHODS FOR HYPERBOLIC...

2021
[11]

A wavelet-free approach for multiresolution-based grid adaptation for conservation laws

Nils Gerhard, Siegfried M¨ uller, and Aleksey Sikstel. A wavelet-free approach for multiresolution-based grid adaptation for conservation laws. Commun. Appl. Math. Comput. , 4(1):108–142, 2022

2022
[12]

A-posteriori error estimates for systems of hyperbolic conservation laws via computing negative norms of local residuals

Jan Giesselmann and Aleksey Sikstel. A-posteriori error estimates for systems of hyperbolic conservation laws via computing negative norms of local residuals. IMA Journal of Numerical Analysis , page drae111, 03 2025

2025
[13]

A priori error estimates of Runge-Kutta discontinuous Galerkin schemes to smooth solutions of fractional conservation laws

Fabio Leotta and Jan Giesselmann. A priori error estimates of Runge-Kutta discontinuous Galerkin schemes to smooth solutions of fractional conservation laws. ESAIM Math. Model. Numer. Anal. , 58(4):1301–1315, 2024

2024
[14]

LeVeque and Blake Temple

Randall J. LeVeque and Blake Temple. Stability of Godunov’s method for a class of 2×2 systems of conservation laws. Trans. Amer. Math. Soc. , 288(1):115–123, 1985

1985
[15]

Convex integration applied to the multi-dimensional compressible Euler equations , volume 2294 of Lecture Notes in Mathematics

Simon Markfelder. Convex integration applied to the multi-dimensional compressible Euler equations , volume 2294 of Lecture Notes in Mathematics . Springer, Cham, [2021] ©2021

2021
[16]

Entropy stability theory for difference approximations of nonlinear conservation laws and related time-dependent problems

Eitan Tadmor. Entropy stability theory for difference approximations of nonlinear conservation laws and related time-dependent problems. Acta Numer., 12:451–512, 2003

2003
[17]

Eleuterio F. Toro. Riemann solvers and numerical methods for fluid dynamics . Springer-Verlag, Berlin, third edition, 2009. A practical introduction. Email address : leotta@mathematik.tu-darmstadt.de

2009

Pith tools

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

[1] [1]

Vanishing viscosity solutions of nonlinear hyperbolic systems

Stefano Bianchini and Alberto Bressan. Vanishing viscosity solutions of nonlinear hyperbolic systems. Ann. of Math. (2) , 161(1):223–342, 2005

work page 2005

[2] [2]

Bressan and H

A. Bressan and H. K. Jenssen. On the convergence of Godunov scheme for nonlinear hyperbolic systems.Chinese Ann. Math. Ser. B , 21(3):269–284, 2000

work page 2000

[3] [3]

Hyperbolic systems of conservation laws , volume 20 of Oxford Lecture Series in Mathematics and its Applications

Alberto Bressan. Hyperbolic systems of conservation laws , volume 20 of Oxford Lecture Series in Mathematics and its Applications . Oxford University Press, Oxford, 2000. The one-dimensional Cauchy problem

work page 2000

[4] [4]

A posteriori error estimates for numerical solutions to hyperbolic conservation laws

Alberto Bressan, Maria Teresa Chiri, and Wen Shen. A posteriori error estimates for numerical solutions to hyperbolic conservation laws. Arch. Ration. Mech. Anal. , 241(1):357–402, 2021

work page 2021

[5] [5]

An instability of the Godunov scheme

Alberto Bressan, Helge Kristian Jenssen, and Paolo Baiti. An instability of the Godunov scheme. Comm. Pure Appl. Math. , 59(11):1604–1638, 2006

work page 2006

[6] [6]

TVB Runge-Kutta local projection discontinuous Galerkin finite element method for conservation laws

Bernardo Cockburn, San Yih Lin, and Chi-Wang Shu. TVB Runge-Kutta local projection discontinuous Galerkin finite element method for conservation laws. III. One-dimensional systems.J. Comput. Phys., 84(1):90–113, 1989

work page 1989

[7] [7]

Multiresolution schemes for conservation laws

Wolfgang Dahmen, Birgit Gottschlich-M¨ uller, and Siegfried M¨ uller. Multiresolution schemes for conservation laws. Numer. Math. , 88(3):399–443, 2001

work page 2001

[8] [8]

Error control for a class of Runge-Kutta discontinuous Galerkin methods for nonlinear conservation laws

Andreas Dedner, Charalambos Makridakis, and Mario Ohlberger. Error control for a class of Runge-Kutta discontinuous Galerkin methods for nonlinear conservation laws. SIAM J. Numer. Anal. , 45(2):514–538, 2007

work page 2007