Evaluation of an eddy-permitting finite-element ocean model in the North Atlantic
A new version of the 3D finite-element primitive-equation ocean model (FEOM) based on tetrahedron partitioning of the computational domain is applied to simulate the North Atlantic circulation at eddy-permitting resolution (1/15°–2°). It relies on a horizontally refined mesh in regions of steep topo...
Gespeichert in:
Veröffentlicht in: | Ocean modelling (Oxford) 2005, Vol.10 (1), p.35-49 |
---|---|
Hauptverfasser: | , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | A new version of the 3D finite-element primitive-equation ocean model (FEOM) based on tetrahedron partitioning of the computational domain is applied to simulate the North Atlantic circulation at eddy-permitting resolution (1/15°–2°). It relies on a horizontally refined mesh in regions of steep topography and allows the sloping bottom to be represented within the
z-coordinate vertical discretization, similar to the so-called shaved-cell approach. It is the first time this approach is used to model large-scale ocean circulation. The FEOM performance in the North Atlantic is compared with that of the finite-difference models of similar resolution of the DYNAMO project. The meridional overturning circulation and heat transport of FEOM agree well with those of the DYNAMO project models, while the mean sea surface height demonstrates the presence of the Gulf Stream recirculation reproduced only by the ISOPYCNIC model of DYNAMO. The annual mean transports of the Gulf Stream and Deep Western Boundary Current at 27°N are of 37
Sv and 17
Sv with core velocities of about 1
m/s and 12
cm/s respectively. Due to flexibility in mesh refinement the FEOM provides a tool for modelling the influence of small-scale phenomena unresolved by current climate models on large-scale ocean circulation. |
---|---|
ISSN: | 1463-5003 1463-5011 |
DOI: | 10.1016/j.ocemod.2004.07.006 |