Improved prediction of coastal storm surge using multi-criteria mesh optimization

Theme: Earth, Atmosphere & Ocean Processes

Primary Supervisor:

Jon French

Geography, UCL

Project Description:

Storm surges are an important agent of coastal flooding that may become more severe as a result of climate change aggravated by accelerated sea-level rise. The complexity of the non-linear hydrodynamic interactions between meteorological and astronomical forcing is an unresolved problem and better modelling is the key to more effective management of storm surge vulnerability, risk, response and recovery. Existing surge prediction models predict inaccurate and unrealistic storm surge water levels in restricted coastal and estuarine waters unless nested grids are used. This project will develop and test a more robust methodology for predicting coastal storm surge water levels that is based on innovative new algorithms that take into account the likely path of the storm surge when generating the computational mesh. The PhD student will:

i) develop a new multi-criteria system to control mesh generation based on multiple functions and datasets, including previous simulation results, to track the storm surge event based on forecast low pressures.

ii) apply optimized meshes to the simulation of historic storm surge events using proven finite element model codes (TELEMAC;; and ADCIRC;

iii) investigate surge-wave coupling, not included in existing operational surge models, and the effect of offshore banks on surge water levels at the coast.

The student will be embedded in environmental modelling groups at UCL and the CASE partner, CEFAS Lowestoft. Training in open-source software, programming, and parallel computing will be provided. This PhD will directly inform the UK Environmental Predictions project (Met Office, NOC, CEH).

Policy Impact of Research:

Improved forecasting of storm surges directly benefits agencies with flood protection responsibilities, operators of key infrastructure (especially ports and power generators) and insurers. A ‘Participatory Modelling’ approach proven in recent NERC projects (NE/J005541/1; NE/M008150/1) will be used to involve stakeholders in the development and testing of the modelling approach and its translation into policy-relevant knowledge.

Stay informed

Subscribe to our RSS newsletter by email.

Find Us

University College London is the administrative lead.

Pearson Building, UCL, Gower Street, London, WC1E 6BT

Follow us on Twitter