- NSF-PAR ID:
- 10035397
- Publisher / Repository:
- Wiley Blackwell (John Wiley & Sons)
- Date Published:
- Journal Name:
- Journal of Geophysical Research: Oceans
- Volume:
- 122
- Issue:
- 5
- ISSN:
- 2169-9275
- Page Range / eLocation ID:
- 3861 to 3883
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
null (Ed.)The energetically independent linear wave and geostrophic (vortex) solutions are shown to be a complete basis for velocity and density variables $(u,v,w,\rho )$ in a rotating non-hydrostatic Boussinesq fluid with arbitrary stratification and non-periodic vertical boundaries. This work extends the familiar wave-vortex decomposition for triply periodic domains with constant stratification. As a consequence of the decomposition, the fluid can be unambiguously separated into decoupled linear wave and geostrophic components at each instant in time, without the need for temporal filtering. The fluid can then be diagnosed for temporal changes in wave and geostrophic coefficients at each unique wavenumber and mode, including those that inevitably occur due to nonlinear interactions. We demonstrate that this methodology can be used to determine which physical interactions cause the transfer of energy between modes by projecting the nonlinear equations of motion onto the wave-vortex basis. In the particular example given, we show that an eddy in geostrophic balance superimposed with inertial oscillations at the surface transfers energy from the inertial oscillations to internal gravity wave modes. This approach can be applied more generally to determine which mechanisms are involved in energy transfers between wave and vortices, including their respective scales. Finally, we show that the nonlinear equations of motion expressed in a wave-vortex basis are computationally efficient for certain problems. In cases where stratification profiles vary strongly with depth, this approach may be an attractive alternative to traditional spectral models for rotating Boussinesq flow.more » « less
-
more » « less
Abstract This paper documents development of a multiple‐Graphics Processing Unit (GPU) version of FUNWAVE‐Total Variation Diminishing (TVD), an open‐source model for solving the fully nonlinear Boussinesq wave equations using a high‐order TVD solver. The numerical schemes of FUNWAVE‐TVD, including Cartesian and spherical coordinates, are rewritten using CUDA Fortran, with inter‐GPU communication facilitated by the Message Passing Interface. Since FUNWAVE‐TVD involves the discretization of high‐order dispersive derivatives, the on‐chip shared memory is utilized to reduce global memory access. To further optimize performance, the batched tridiagonal solver is scheduled simultaneously in multiple‐GPU streams, which can reduce the GPU execution time by 20–30%. The GPU version is validated through a benchmark test for wave runup on a complex shoreline geometry, as well as a basin‐scale tsunami simulation of the 2011 Tohoku‐oki event. Efficiency evaluation shows that, in comparison with the CPU version running at a 36‐core HPC node, speedup ratios of 4–7 and above 10 can be observed for single‐ and double‐GPU runs, respectively. The performance metrics of multiple‐GPU implementation needs to be further evaluated when appropriate.
-
more » « less
Summary We develop one‐way coupling methods between a Boussinesq‐type wave model based on the discontinuous Galerkin finite element method and a free‐surface flow model based on a mesh‐free particle method to strike a balance between accuracy and computational cost. In our proposed model, computation of the wave model in the global domain is conducted first, and the nonconstant velocity profiles in the vertical direction are reproduced by using its results. Computation of the free‐surface flow is performed in a local domain included within the global domain with interface boundaries that move along the reproduced velocity field in a Lagrangian fashion. To represent the moving interfaces, we used a polygon wall boundary model for mesh‐free particle methods. Verification and validation tests of our proposed model are performed, and results obtained by the model are compared with theoretical values and experimental results to show its accuracy and applicability.
