On Affordable High-Order Entropy-Conservative/Stable and Well-Balanced Methods for Nonconservative Hyperbolic Systems

Marco Artiano, Hendrik Ranocha · 2026

Read On Affordable High-Order Entropy-Conservative/Stable and Well-Balanced Methods for Nonconservative Hyperbolic Systems by Marco Artiano, Hendrik Ranocha (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-18978.txt.

Introduction

Artiano and Ranocha build cheaper high-order methods that keep entropy steady for wave-like systems. Past methods rest on pair fluxes and special grid operators. Their work gives simple forms for the extra terms that nonconservative systems need. They work out when such terms keep entropy in a three-point setup. The forms extend to many dimensions on straight and curved grids. Balance traits for steady states carry over as well. They use the recipe to derive several new schemes.

Catalog record

CREATORMarco Artiano, Hendrik Ranocha
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE167170
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.18978

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

On Affordable High-Order Entropy-Conservative/Stable and Well-Balanced Methods for Nonconservative Hyperbolic Systems Marco Artiano, Hendrik Ranocha Abstract Many entropy-conservative and entropy-stable (summarized as entropy-preserving) methods for hyperbolic conservation laws rely on Tadmor’s theory for two-point entropy-preserving numerical fluxes and its higher-order extension via flux differencing using summation-by-parts (SBP) operators, e.g., in discontinuous Galerkin spectral element methods (DGSEMs). The underlying two-point formulations have been extended to nonconservative systems using fluctuations by Castro et al. (2013, doi:10.1137/110845379 ) with follow-up generalizations to SBP methods. We propose specific forms of entropy-preserving fluctuations for nonconservative hyperbolic systems that are simple to interpret and allow an algorithmic construction of entropy-preserving methods. We analyze necessary and sufficient conditions, and obtain a full characterization of entropy-preserving three-point methods within the finite volume framework. This formulation is extended to SBP methods in multiple space dimensions on Cartesian and curvilinear meshes. Additional properties such as well-balancedness extend naturally from the underlying finite volume method to the SBP framework. We use the algorithmic construction enabled by the chosen formulation to derive several new entropy-preserving schemes for nonconservative hyperbolic systems, e.g., the compressible Euler equations of an ideal gas using the internal energy equation and a dispersive shallow-water model. Numerical experiments show the robustness and accuracy of the proposed schemes. keywords structure-preserving methods, entropy-conserving methods, entropy-stable methods, kinetic energy-preserving methods, pressure equilibrium-preserving methods, nonconservative terms, summation-by-parts operators, discontinuous Galerkin methods, curvilinear coordinates AMS subject classification. 65M12, 65M20, 65M70, 65M60, 65M06, 65M08 1 Introduction For hyperbolic conservation laws such as \partial_{t}\bm{u}+\partial_{x}\bm{f}(\bm{u})=0, (1) convex entropies are crucial for analysis and numerics. From a theoretical point of view, they provide stability for classical solutions [ 27 , Section V] as well as selection criteria for weak [ 27 , Sections VIII and IX] or even more generalized solutions [ 34 ] . From a numerical point of view, entropy-stable methods improve the robustness of under-resolved simulations significantly [ 42 , 18 , 90 , 84 , 7 , 4 ] . Today, many efficient and entropy-conservative/stable (EC/ES, summarized as entropy-preserving) methods for conservation laws [ 74 , 107 ] are based on the seminal work of Tadmor [ 94 , 95 ] , who introduced the concept of entropy-preserving numerical fluxes for finite volume methods \partial_{t}\bm{u}_{i}+\frac{1}{\Delta x_{i}}\Bigl(\bm{f}^{\mathrm{num}}(\bm{u}_{i},\bm{u}_{i+1})-\bm{f}^{\mathrm{num}}(\bm{u}_{i-1},\bm{u}_{i})\Bigr)=\bm{0}. (2) The extension to high-order methods has been successively obtained in [ 58 , 37 , 24 , 70 , 19 , 20 ] , relying on the general framework of summation-by-parts (SBP) operators [ 93 , 35 ] . SBP operators are discrete derivative and integration operators that mimic integration by parts. While they have originally been developed for finite differences (FDs) [ 55 , 92 , 15 ] , their general framework also includes finite volumes [ 64 ] , continuous Galerkin (CG) finite elements [ 44 , 45 , 2 ] , discontinuous Galerkin (DG) methods [ 43 , 14 ] , flux reconstruction (FR) [ 47 , 101 , 73 ] , active flux methods [ 33 , 8 ] , meshless schemes [ 46 , 57 ] , and cut-cell methods [ 66 , 96 ] . In contrast to classical conservation laws such as ( 1 ), nonconservative hyperbolic systems \partial_{t}\bm{u}+\bm{A}(\bm{u})\partial_{x}\bm{u}=\bm{0} (3) are more difficult to handle. For example, the nonconservative product needs to be defined carefully for weak solutions; the shock relations depend not only on the solution states but also on the path connecting these states [ 62 ] . This has led to the development of path-conservative numerical methods for nonconservative systems [ 65 ] ; extensions to DG methods have been carried out by various authors [ 82 , 39 ] . However, path conservation alone is not enough to ensure correct solutions [ 17 , 1 ] . Indeed, nonconservative systems are susceptible to the correct form of (artificial, effective) dissipation. Thus, entropy-preserving schemes are important here as well [ 16 , 9 ] . The extension of Tadmor’s entropy framework to nonconservative systems has been initiated in [ 16 ] by considering finite volume methods in fluctuation form, i.e., \partial_{t}\bm{u}_{i}+\frac{1}{\Delta x_{i}}\Bigl(\bm{D}^{+}(\bm{u}_{i-1},\bm{u}_{i})+\bm{D}^{-}(\bm{u}_{i},\bm{u}_{i+1})\Bigr)=\bm{0}, (4) where the fluctuations satisfy the consistency condition \bm{D}^{\pm}(\bm{u},\bm{u})=\bm{0} and an additional path consistency condition involving an integral [ 65 , 16 ] . Assuming that the nonconservative system ( 3 ) is equipped with a convex entropy U , i.e., the entropy inequality \partial_{t}U(\bm{u})+\partial_{x}F(\bm{u})\leq 0 (5) is demanded for weak solutions and holds as equality for smooth solutions, Castro et al. [ 16 , Theorem 2.1] derived the sufficient condition \forall\bm{u}_{-},\bm{u}_{+}\colon\quad\bm{\omega}(\bm{u}_{-})\cdot\bm{D}^{-}(\bm{u}_{-},\bm{u}_{+})+\bm{\omega}(\bm{u}_{+})\cdot\bm{D}^{+}(\bm{u}_{-},\bm{u}_{+})=F(\bm{u}_{+})-F(\bm{u}_{-}) (6) for an EC method, where \bm{\omega}=U^{\prime}(\bm{u}) are the entropy variables. These methods have been extended to 1D SBP DG methods by Renac [ 81 , 26 ] and to curved meshes in multiple dimensions by Waruszewski et al. [ 104 ] , who also considered nonconservative terms of the form \bm{H}(\bm{u})\partial_{x}\bm{g}(\bm{u}) instead of \bm{A}(\bm{u})\partial_{x}\bm{u} . Recently, Ersing and Winters [ 31 ] proposed two particular strategies to compute EC fluctuations: one based on an integral formulation and another assuming a specific form of two-point fluxes, provided that a closed-form expression can be found. However, their construction still relies on integrals. Tadmor presented a general choice of EC fluxes for conservation laws [ 94 , 95 ] based on an integral form. In practice, these integrals are typically hard or even impossible to evaluate (exactly and efficiently). Thus, it took some time until affordable EC fluxes have been constructed without the need to compute integrals [ 48 , 23 , 70 ] . Moreover, Castro et al. [ 16 , Remark 4.2] observed: This example indicates that the choice of paths is not crucial in determining which solutions are approximated by the scheme. Instead, the numerical viscosity operator (that matches with the underlying viscosity) decides which weak solution the scheme will converge to. This is fortunate, since the correct choice of paths is often far from trivial, and many works have thus resorted to the default choice of a straight line path. This motivates our current manuscript. By focusing on the entropy, we first propose concrete forms of fluctuations that are easy to interpret and amenable to the algorithmic construction of affordable entropy-conserving methods, without the need to evaluate any integrals (including integrals related to choosing a path). We provide both necessary and sufficient conditions for entropy conservation and stability, and demonstrate how to use them to construct entropy-preserving methods for several hyperbolic systems. Moreover, we extend the results to high-order SBP methods on curved meshes in multiple dimensions. Furthermore, we demonstrate how well-balanced three-point finite volume schemes extend to well-balanced high-order SBP methods on curved meshes. In contrast to works on path-conservative methods for nonconservative systems, we consider the general form \partial_{t}\bm{u}+\partial_{x}\bm{f}(\bm{u})+\bm{H}(\bm{u})\partial_{x}\bm{g}(\bm{u})=\bm{0}, (7) separating conservative parts from nonconservative ones and allowing general derivatives in nonconservative terms. This is important for many applications, e.g., to allow differentiating the pressure or the velocity for the compressible Euler equations, which are not conserved variables. Moreover, we do not only provide sufficient conditions, but also study their necessity in a more general context without requiring convexity of the entropy. In fact, while Tadmor’s integral form of an EC flux requires a convex entropy (to obtain the entropy variables \bm{\omega}=U^{\prime}(\bm{u}) as a bijective mapping from the conserved

Read the complete text