A 2.5D Model for the Equations of the Ocean and the Atmosphere

Abstract

The primitive equations (PEs) of the atmosphere and the ocean without viscosity are considered. A 2.5D model is introduced, whose motivation is described in the Introduction. A set of nonlocal boundary conditions is proposed, and well-posedness is established for the flows linearized around a constant velocity stratified flow; homogeneous and nonhomogeneous boundary conditions are considered. A related model of dimension 2.5, of physical interest but with fewer degrees of freedom, is also considered at the end.

Numerical approximation of the inviscid 3D primitive equations in a limited domain

“…The well-posedness result for the linearized equations was established in [2], and the numerical simulations of the nonlinear equations on a nested set of domains were discussed in [3].…”

A technique for non-invasive application-level checkpointing

“…Scientific applications that take enormous amount of time to execute (e.g., simulation for protein structure prediction [1] or climate modeling [2]) and are run in distributed, dynamic and heterogeneous environments, like a grid, can benefit considerably from checkpointing. In case of failures or changes in the availability of underlying resources, instead of restarting the application from the beginning, the application is restarted from the latest checkpoint.…”

Simulations of the 2.5D inviscid primitive equations in a limited domain

“…This article is concerned with the primitive equations (PEs) without viscosity in space dimension 2.5, and is related to the more theoretical article [3].…”

“…In this way we were led to choose, for the y-direction, a three-mode linear finite element model. In the article [3], we presented the full derivation of the model and studied the wellposedness of the linearized equations. The present article is aimed at actually testing in a physically relevant context the model in [3] which we here call a 2.5D model.…”

“…In the article [3], we presented the full derivation of the model and studied the wellposedness of the linearized equations. The present article is aimed at actually testing in a physically relevant context the model in [3] which we here call a 2.5D model. In fact the model presented in [3] was a linear (linearized) one.…”