A Framework for Running the ADCIRC Discontinuous Galerkin Storm Surge Model on a GPU

DuChene, Michael; Spagnuolo, Anna Maria; Kubatko, Ethan J.; Westerink, Joannes J.; Dawson, Clint

doi:10.1016/j.procs.2011.04.220

Cited by 4 publications

(1 citation statement)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…to support the use of high-order interpolants in data-rich settings. Additionally, the methods have been shown to be highly parallelizable in implementations on both traditional parallel platforms [4,5] and graphics-processing units (GPUs) [6,7,8]. And finally, DG methods possess the important property of local (element-wise) conservation, which is of fundamental importance in hydrological modeling, especially when coupling various model components [2,9].…”

Section: Introductionmentioning

confidence: 99%

A multidimensional discontinuous Galerkin modeling framework for overland flow and channel routing

West

Kubatko

Conroy

et al. 2017

Advances in Water Resources

View full text Add to dashboard Cite

In this paper, we present the development and application of a new multidimensional, unstructured-mesh model for simulating coupled overland/openchannel flows in the kinematic wave approximation regime. The modeling approach makes use of discontinuous Galerkin (DG) finite element spatial discretizations of variable polynomial degree p, paired with explicit Runge-Kutta time steppers, and is supported by advancements made to an automatic mesh generation tool, Admesh+, that is used to construct constrained triangulations for channel routing. The developed modeling framework is applied to a set of four test cases, where numerical results are found to compare well with known analytic solutions, experimental data and results from another well-established (structured, finite difference) model within the area of application. The numerical results obtained demonstrate the accuracy and robustness of the developed modeling framework and highlight the potential benefits of using p (polynomial) refinement for hydrological simulations.

show abstract