Modelling the influence of small-scale effects upon the larger scale: an oceanographic challenge

Abstract

The problem of resolving or parameterising small-scale processes in oceanographic models and the extent to which small-scale effects influence the large scale are briefly discussed and illustrated for a number of cases. For tides and surges in near-shore regions, the advantages of using a graded mesh to resolve coastal and estuarine small-scale features are demonstrated in terms of a west coast of Britain unstructured mesh model. The effect of mesh resolution upon the accuracy of the overall solution is illustrated in terms of a finite element model of the Irish Sea and Mersey estuary. For baroclinic motion at high Froude number, the effect of resolving small-scale topography within a non-hydrostatic model is illustrated in terms of tidally induced mixing at a single sill, or two closely spaced sills. The question of how to parameterise small-scale nonlinear interaction processes that lead to significant mixing, in a form suitable for coarser grid hydrostatic models, is briefly considered. In addition, the importance of topographically induced mixing that occurs in the oceanic lateral boundary layer, namely, the shelf edge upon the large-scale ocean circulation is discussed together with the implications for coarse grid oceanic climate models. The use of unstructured grids in these models to enhance resolution in shelf-edge regions in a similar manner to that used in storm surge models to enhance near coastal resolution is suggested as a suitable "way forward" in large-scale ocean circulation modelling.

The Dispersion Algorithm: Consideration of Bathymetry in 2D Mesh generation in Ocean Modeling

“…After presenting these factors, the authors of this inspiring paper attempted to qualify and quantify the influence of each factor on the horizontal mesh resolution needed, for a more accurate modeling of the oceans. In recent years, the concept of relating the elements sizes to different physical factors has been widely used by many researchers [7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23]. Among the most important ones are works done by Robin et al, Jones, Horsburgh, Wolf, Parrish et al and White et al Robins et al [7,8], used this concept to simulate the water flows and sediments transport in coastal models.…”