New design floods for a complex, changing and increasingly interconnected reality

Infrastructure is commonly designed to safely accommodate design events with a specified return period such as, for example, the 100-year event. Similarly, the 100-year flood is used to demarcate the flood risk zone along river corridors where construction of homes and businesses is not permitted. However, recent extreme and record-breaking flood events in the north of England and Scotland in late 2015 (and countless events before that) have once again highlighted the vulnerability of homes, businesses and critical infrastructure to flood risk. Given the apparent increasing frequency of flooding and the existing infrastructure at risk, traditional engineering design criteria based on simple and stationary return periods need to be updated to better reflect the evolving reality in which the infrastructure operates. In particular, a new approach needs to better incorporate external drivers of flood risk to enable assessment of impact of climate and land-use change on flood risk. The new models must also consider flooding on a wider spatial scale than is currently done, e.g. quantifying the reliability of a railway network in its entirety rather than a point by point assessment, which is likely to underestimate the true risk. The main aim of the research to be undertaken in this project is therefore to develop the scientific foundation for a new probabilistic framework for risk assessment of complex infrastructure systems in a non-stationary environment, replacing the increasingly outdated concept of return periods. In particular, the research will develop new hydrological risk-models models for:
• establishing tele-connections between large-scale climatic drivers (e.g. El Nino) and local scale flooding
• quantify the impact of urbanisation across scales (local to basin scale) on changing flood risk
• develop multivariate extreme value models accounting for the dependence between different type of flood events (pluvial and fluvial) and flooding at different locations

The project will build on existing national databases of hydro-meteorological flood and rainfall events, as well as large-scale climatological data from regional and global climate models. The expected outcome is a new set of hydrological design models with high potential for implementation in flood risk guidelines in the UK and beyond. The project is an ambitious undertaking which will provide a motivated student with opportunities to interface with several scientific disciplines, including: civil engineering, meteorology/climatology, mathematics and statistics, and geography as well as exposure to government and industry organisations involved in flood risk management.

Applicants must have a keen interest in urban water and energy issues with a good knowledge of mathematical and statistical analysis and computer programming (e.g. Python/R). Applicants should have a degree in a relevant engineering, physics or applied mathematics related discipline as well as strong English communication skills, both verbally and written.