Open Access
Application of an unstructured mesh model to the determination of the baroclinic circulation of the Irish Sea
Author(s) -
Xing Jiuxing,
Davies Alan M.,
Jones J. Eric
Publication year - 2011
Publication title -
journal of geophysical research: oceans
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.67
H-Index - 298
eISSN - 2156-2202
pISSN - 0148-0227
DOI - 10.1029/2011jc007063
Subject(s) - baroclinity , geology , finite volume method , forcing (mathematics) , circulation (fluid dynamics) , polygon mesh , climatology , atmospheric sciences , mechanics , geometry , physics , mathematics
A three‐dimensional finite volume model with horizontally varying, but fixed in time mesh, is used to compute the baroclinic circulation of the Irish Sea during 1995. Tidal forcing was applied along the model's open boundary with meteorological forcing taken from observations. Initial calculations were performed with a mesh that had high resolution in the well mixed near coastal region; a necessary requirement in order to reproduce tides in the region, although offshore in the stratified area the mesh was slightly coarser than that used in earlier finite difference models. Subsequent calculations were performed using an enhanced resolution which is significantly finer than earlier finite difference models in the offshore region which is thermally stratified in summer due to solar heating and low tidal mixing. This produces a cold water bottom dome separated from the well mixed shallow water regions by strong tidal fronts. Calculations show that both model meshes can reproduce the observed major features of the baroclinic circulation of the western Irish Sea, with the coarse mesh model giving comparable results to earlier finite difference models. In the case of the finer mesh model there are sharper horizontal density gradients in the region of the fronts, which show the presence of baroclinic instability and associated small scale variability as observed in satellite images but not found in the coarser mesh model due to lack of resolution. Results from the fine mesh model show significantly more spatial variability comparable to that found in the measurements.