The University of Sheffield
Department of Civil and Structural Engineering

Integrated Numerical Modelling Of Multi-layered Tsunami Waves

Supervisor: Dr Georges Kesserwani

Joint Supervisor: Dr Shao Songdong

The conventional shallow water equations (single layer) (under hydrostatic pressure assumption) have long been used to produce computer codes for tsunami modelling and forecasting. However, a tsunami wave involves movement of the entire water column from surface to seafloor, across which multi-layered processes and different densities occur. Within the scope of a depth-integrated approach, the classical Boussinesq formulation has been used to account for the non-hydrostatic pressure components and introduce a vertical velocity in response to it. The Boussinesq approach is more accurate due to the increased level of physical complexity added to the shallow water equations. However, it consumes huge computer resources and adds numerical challenges to handle extra dispersion terms. The class of multi-layer shallow water models has been developed to handle flow problems involving superposition of immiscible shallow fluids without the use of empirical relations for energy dissipation and provide comparable results with those of extended Boussinesq models. Stability, however, is a critical issue, with increasing difficulties in proportion with the number of flow layers involved. The resulting errors in flux estimations often become the source of instability. In the context of a three-layer interaction: shallow layer, intermediate layer and deep layer; this PhD work aims to build a tsunami computer model based on both (1) the Boussinesq approach and (2) the multi-layer approach. The project will augment the development of an (available) finite volume 2D numerical model to meet objectives (1) and (2), and then explore and compare the performance of both models on hypothetical and experimental test cases. Result analysis will be performed to identify whether the three-layer approach can have significant practical implications on coastal flood risk management.

Suitable applicants are those with any Engineering, or Mathematics and Physics backgrounds. Experience in Computational Fluid Mechanics, GIS and in programming languages is a plus.

This project is NOT FUNDED, although Departmental/University scholarships are available for applicants who can demonstrate strong evidence of research potential.