High-Order Discontinuous Galerkin Methods for Complex Physics

60 concepts48 papers1 proposal

Numerical simulation often forces a choice between two worlds: the robust, locally conservative nature of Finite Volume (FV) methods and the high-order spatial accuracy of Finite Element Methods (FEM). The Discontinuous Galerkin (DG) method bridges this gap, offering a framework that is both high-order accurate and inherently conservative.

The Core Intuition: Localized Polynomials

In traditional finite element methods, the solution is forced to be continuous across the boundaries of computational elements. While this works well for smooth problems, it struggles with "sharp" physics like shock waves or steep gradients.

As Hailiang Liu and Seshu Tirupathi et al. describe, the DG method breaks this continuity requirement. It partitions the domain into cells and approximates the solution within each cell using independent piecewise polynomials. Because these polynomials are not tied to their neighbors, the solution can "jump" at the interfaces.

This localized representation provides two immediate advantages:

  1. High-Order Accuracy: By increasing the degree of the polynomial within a cell ($p$-refinement), researchers can achieve very high accuracy without necessarily making the mesh smaller ($h$-refinement), as explored by A. Bolis et al..
  2. Robustness to Discontinuities: The method naturally handles sharp features like shocks in plasma physics George Em Karniadakis or turbidity currents S. Parkinson et al. because the mathematical framework expects and manages jumps at element boundaries.

Communication via Numerical Fluxes

If every cell is independent, how does the global physics emerge? The answer lies in numerical fluxes. At the interface between two cells, the DG method uses a flux function—often a Riemann solver—to communicate information.

This mechanism ensures that the method remains locally conservative, a property Stephan C. Kramer et al. highlight as critical for coastal ocean modeling. By treating the interfaces with specialized fluxes, the method can maintain stability in advection-dominated flows where traditional methods might produce unphysical oscillations. To further ensure physical reality, such as preventing negative pressure or concentration, authors like Hailiang Liu and Forrest W. Glines et al. employ positivity-preserving limiters that strictly enforce physical bounds.

Flexibility Across Geometries and Scales

Because DG elements only communicate through their faces, the method is exceptionally flexible. Tim Warburton et al. demonstrate that DG can be applied to hybrid grids—mixtures of structured and unstructured elements—allowing it to conform to complex engineering geometries, such as aircraft engine mounts Daniel Lindblad et al..

This local independence also makes DG highly parallelizable. Forrest W. Glines et al. note that the high "arithmetic intensity" (the ratio of math operations to data movement) makes DG particularly well-suited for modern GPU architectures.

Specialized Variants for Diffusion and Multiscale Problems

While standard DG is ideal for advection (transport), handling diffusion (second-order derivatives) requires care because the solution is discontinuous. Several specialized frameworks have emerged to address this:

  • Local Discontinuous Galerkin (LDG): This approach, used by Francis Filbet for surface deformation, rewrites higher-order equations as a system of first-order equations.
  • Hybridized Discontinuous Galerkin (HDG): To reduce computational cost, Yalchin Efendiev utilizes HDG to solve multiscale problems in porous media. HDG introduces new variables only on the element boundaries, significantly reducing the number of global unknowns.
  • Weak Galerkin (WG): As proposed by Chunmei Wang et al., this method uses generalized distributions to interpret differential operators, allowing for even simpler basis functions while maintaining accuracy for complex fourth-order problems like fluorescence tomography.

By combining the geometric flexibility of unstructured meshes with the high-order precision of spectral methods, Discontinuous Galerkin formulations provide a rigorous, scalable foundation for simulating the most challenging phenomena in fluid dynamics, wave propagation, and beyond.

Key Foundational Work

A discontinuous Galerkin method for two-temperature plasmas (2005) — contributed 2 concept elements to this direction.
A discontinuous Galerkin spectral/hp method on hybrid grids (2000) — contributed 2 concept elements to this direction.
History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_discontinuous_galerkin_method_dfd36790,
  title = {Discontinuous Galerkin method},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/discontinuous-galerkin-method/rdr_dfd3679049dd4c4e96d1951eafb40019},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}

This exact version (version 1, frozen text):

@misc{lacuna_discontinuous_galerkin_method_dfd36790_v1,
  title = {Discontinuous Galerkin method},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/discontinuous-galerkin-method/rdr_dfd3679049dd4c4e96d1951eafb40019/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake