The Mathematics of Planet Earth – EPSRC Centre for Doctoral Training (MPE-CDT), together with the Smith Institute and EDF Energy, has one Industrial CASE-EPSRC studentship available to start in October 2018. The four-year PhD project, “Building physical constraints into weather-related risk estimates”, will aim at the development of robust and efficient estimation methods for very high quantiles or for the occurrence probability of a particular type of extreme event.
When required to extrapolate to very rare return periods, standard extreme value models and their associated statistical inference techniques can lead to estimates with a large amount of uncertainty. The estimation of one specific parameter, the shape parameter (or the extreme value index), can suggest an unbounded growth in the hazard curve, that is, for rarer return periods the respective return level can become exceedingly large; this sort of behaviour is often found for rainfall datasets. In many situations, these estimates start to exceed plausible levels. Such large values tend to reduce the confidence that end-users have in the final results from extreme value models.
In the context of recent extreme weather events (for example, Hurricane Matthew and Storm Ophelia), natural hazard characterization continues to be an important area of research for EDF Energy. Developments in statistical approaches, climate modelling and computational power continue to drive research in this field, and this in turn drives EDF Energy to undertake further research. This PhD project is seen as a way to improve the natural hazard characterization approaches used within EDF Energy. The research work will encompass both the physical and the extreme value statistical frameworks. The overall goal is to obtain more reliable constraints on extreme rainfall risk from historical records, using physical principles to draw inferences from observable causes to certain extreme events. In order to accurately estimate the risk of extreme weather events, this project will combine new developments in extreme value theory with recent advances in approaches for understanding the physical drivers of extreme events, where probabilities will be estimated using a storylines construction in a physically coherent way.
The ICASE PhD project will be based at the University of Reading. The work will be jointly supervised by Dr Claudia Neves (Dept. Mathematics and Statistics), Prof Ted Shepherd (Dept. of Meteorology) and Dr Hugo Winter (EDF Energy). Through the affiliation with the MPE-CDT, the successful candidate will have the opportunity to attend core CDT courses and to participate in MPE regular events and personal development activities. The prospective PhD student will also engage in bespoke training in group interaction, communication, and presentation skills. The direct contact with EDF Energy will help to provide experience of working in an industry context and presenting work to a wide range of end-users.
We are looking for highly motivated candidates with knowledge in data and time series analysis, and experience in technical and/or statistical programming (Matlab, Python, R or a similar language). Some knowledge of the physical principles behind weather and climate would be an asset.
How to apply:
Applications should be lodged via the MPECDT applications procedure, indicating that the candidate wishes to be considered for this joint project. The deadline for applications is 31 January 2018, and interviews will be held on 21 February.