Monday 20 Jan 2014: Global modelling of ocean tides (Taylor and Francis sponsored)
Stephen Griffiths - University of Leeds
Harrison 101 15:00-16:00
A longstanding challenge in oceanography is the production of accurate global maps of the amplitude and phase of the ocean tides, from simple hydrodynamical equations with appropriate astronomical forcing and global topography. Over the last forty years, such prognostic models have improved in accuracy, partly due to increases in computing power, but also as the role of supposedly secondary physical processes have been recognised. Here, the development of a new prognostic global tidal model is described, based upon the solution of large sparse matrix systems coupled with iterations to account for weak nonlinear processes. The resulting solutions are rather sensitive to certain model parameters, so two simplified theoretical models of shallow-water waves (in Cartesian geometry) are developed to understand this sensitivity. The first model is of waves allowing for self-gravitation of the water column and the presence of an elastic medium beneath, the combined effects of which can retard waves by up to 10%. The second model is of wave propagation along coastlines as resolved by finite-difference schemes, illustrating the importance of accurate numerical implementation of (coastal) boundary conditions.