Key Words: Compound events; Multi-hazards; Industry application; Multivariate extreme value theory.Abstract:Resilience of pre-existing and new-build infrastructure to natural hazards is of key interest for many different industries (e.g. energy, water, transport). In most situations, studies analyzing the risk posed by single natural hazards have already been undertaken and relevant protection measures have been implemented. However, when considering the potential impacts of compound events or multi-hazards there can be less confidence in which combinations need to be considered and how to estimate the risks associated to these multi-hazard scenarios. Certain industries (e.g. nuclear) have already undertaken several projects on the occurrence and risks posed by multi-hazards (e.g. ASAMPSA-E, NARSIS), whereas other industries are still trying to understand their risks and which questions need to be posed. The EDF Energy R&D UK Centre are part of an industry scheme funded by NERC called the Environmental Risks for Infrastructure Innovation Programme (ERIIP) which aims to connect academics to industrial organisations and undertake translational research. One of the key topics of further research identified by this group is the topic of compound events and multi-hazards. A recent review of knowledge on multi-hazards was undertaken by the British Geological Survey (BGS) which highlighted the state of knowledge across UK infrastructure owners. This presentation will start by summarizing this review to pull out some key themes for future research in this area. Then, two different ongoing research projects will be outlined which look to address the key themes coming out of the review. One project is attempting to better understand the different multivariate statistical methods that are available for assessing the probability of multi-hazards. The application of the different models outlined in this work will be shown on an example of extreme precipitation and wind speed. The other project aims to better understand the overarching meteorological conditions that can lead to compound flooding at coastal sites around the UK. This focuses less on estimating joint probabilities, but more on producing clear visualisations for end-users.
