The Teignmouth Model: Validation and evaluation of Delft3D-MOR with COAST3D Pilot campaign data

Computer models are commonly used to simulate the behaviour of the coast in response to natural processes (e.g. storms, extreme sea levels) or management plans (e.g. beach nourishment, channel dredging). The COAST3D project was initiated to improve and validate these numerical models. COAST3D stands for Coastal Study of ThreeDimensional Sand Transport Processes and Morphodynamics. In March 1999, a COAST3D Pilot experiment was held in Teignmouth on the coast of Devon in Southwest England.Teignmouth has a very irregular coastline. It comprises a tidal inlet adjacent to the beach and a sandstone cliff, making it three-dimensional.This study involves the modelling of the Teignmouth site with data from the Pilot campaign. The general goal is to validate Delft3D-MOR against the field data taken in Teignmouth. For this purpose, a model grid covering the Teignmouth site is constructed. The hydrodynamic boundary conditions are determined by nesting the Teignmouth gridinto a larger model, the Lyme Bay Model, which is in turn nested into the Continental Shelf Model. After the nesting procedure, the boundary conditions are calibrated with the use of water levels recorded during the Pilot campaign.The influence of the Teignmouth estuary is calibrated by varying the bed roughness inside the estuary. However, the tidal flow through the estuary mouth can not be modelled accurately, as the bathymetry data used in the Teignmouth model is outdated. The different COAST3D modelling teams agreed to carry out three common test cases: 1) A spring tidal cycle without waves. 2) A hypothetical situation with high waves and a fixed water level. 3) A neap tidal cycle with large waves.The focus of this study is also on these three test cases.The suspended and bedload transports and resulting bed-level changes are also computed for each test case. This is done with and without intratidal bed updating.