 |
| Titel |
The application of ICOM, a non-hydrostatic, fully unstructured mesh model in large scale ocean domains |
| VerfasserIn |
Stephan C. Kramer, Matthew D. Piggott, Colin J. Cotter, Chris C. Pain, Rhodri B. Nelson |
| Konferenz |
EGU General Assembly 2010
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 12 (2010) |
| Datensatznummer |
250041275
|
|
|
|
|
|
| Zusammenfassung |
| There are many apparent advantages of the application of unstructured meshes in ocean
modelling: a much better representation of the coastal boundaries, the ability to focus
resolution in areas of interest, or areas of intensified flow, such as boundary currents, etc.
In particular with adaptive mesh technology, where the mesh is adapted during
the simulation as the flow evolves, one is able to resolve much smaller features
in the often turbulent ocean flow, than would be possible with fixed, structured
mesh models. The Imperial College Ocean Model[1], is a non-hydrostatic ocean
model that employs fully unstructured adaptive meshes, that allow focussing of
resolution not only in the horizontal but also in the vertical. This enables the modelling
of physical processes, such as open ocean deep convection, density driven flows
on a steep bottom topography, etc. that are very important for the global ocean
circulation.
The Imperial College Ocean Model has been applied succesfully in the modelling of
many of these processes. On the other hand hydrostatic, layered ocean models have a
significant advantage in large areas of the oceans where the hydrostatic assumption is valid.
The fact that with fully unstructured meshes it is no longer straightforward to separate
horizontal, baroptropic modes and vertical, baroclinic dynamics, has consequences for both
numerical accuracy and the efficiency of the linear solvers. It has therefore been a
challenge for ICOM to remain competitive in these areas with layered mesh models.
These problems have been overcome by, amongst others, the development of a new
mesh adaptation technique that maintains a columnar structure of the mesh in such
areas. The application of multigrid techniques has improved the effiency of the
non-hydrostatic pressure solve[2] in such a way that convergence is now independent of
aspect ratio, which makes the pressure solve competitive with that of a hydrostatic
model.
In this contribution an overview will be given of some of the difficulties that
were encountered in the application of ICOM in large scale, high aspect ratio ocean
domains and how they have been overcome. A large scale application in the form of a
baroclinic, wind-driven double gyre will be presented and the results are compared to
two other models, the MIT general circulation model (MITgcm, [3]) and NEMO
(Nucleus for European Modelling of the Ocean, [4]). Also a comparison of the
performance and parallel scaling of the models on a supercomputing platform will be
made.
References
[1]Â Â Â M.D. Piggott, G.J. Gorman, C.C. Pain, P.A. Allison, A.S. Candy, B.T.
Martin and W.R. Wells, "A new computational framework for multi-scale ocean
modelling based on adapting unstructured meshes", International Journal for
Numerical Methods in Fluids 56, pp 1003 - 1015, 2008
[2]Â Â Â S.C. Kramer, C.J. Cotter and C.C. Pain, "Solving the Poisson equation on
small aspect ratio domains using unstructured meshes", submitted to Ocean
Modelling
[3]Â Â Â J. Marshall, C. Hill, L. Perelman, and A. Adcroft, "Hydrostatic,
quasi-hydrostatic, and nonhydrostatic ocean modeling", J. Geophysical Res.,
102(C3), pp 5733-5752, 1997
[4]Â Â Â G. Madec, "NEMO ocean engine", Note du Pole de modélisation, Institut
Pierre-Simon Laplace (IPSL), France, No 27 ISSN No 1288-1619 |
|
|
|
|
|
|