What is the UK’s present day storm surge risk, and how will this change in the future? Wind-driven coastal flooding or ‘Storm Surge’ is an important risk to the UK, but its standard assessment methodology (SRJPM) dates to 1995 and does not include spatial correlation between assets at risk (e.g. houses and infrastructure; Fig. 1). This project will address that limitation, and progress this research field by: i) taking advantage of the developments in statistical techniques and climate modelling that includes greenhouse gas forcing scenarios; and ii) integrating this with advanced numerical modelling of storm surge and coastal flood inundation, and risk analysis [Royse et al., 2014; Befort et al., 2015; Yin et al., 2016].
Excitingly, modelling will go all the way from surges driven by off-shore storms to inland coastal inundation (i.e. flooding) and risk assessment. It aims to quantify all likely surge disasters, focusing on the national (i.e. whole of UK) and but illustrated with local scale (i.e. urban centre) case studies.
Risk assessment will use ‘catastrophe modelling’, a GIS methodology from the (re-)insurance industry that links physical models of hazards to financial loses; effectively ‘Monte Carlo’ simulation. It uses many (e.g., 10,000) events and their damage ‘footprints’ to simplify risk modelling of complex systems. It allows a probabilistic assessment of losses in order to plan ahead and minimise likely worst-case scenarios, but is relatively under-used in academia giving the potential for critical insights (e.g. the spatial correlation of impacts) into present and future coastal risk in the UK.
A core of the work is low risk, but scope exists for a student to innovate and excel (e.g. focus on critical infrastructure or ‘indirect impacts’). Specific objectives are:
O1 – Establishing a robust, spatially varying and probabilistic set of wind-driven coastal (i.e. skew surge) sea height conditions for extreme events. This will be done by comparing a multivariate extreme value statistical approach [Lamb et al., 2010] with expectations from historical data using the approach of Befort et al. . Storm surge modelling (e.g. with ADCIRC) will fill the gaps between tide gauges.
O2 – Conducting coastal flood inundation modelling, at the national scale at coarse (100m) and finer spatial resolution (10m) in 5-6 urban centres vulnerable to storm surge. Future sea level projections and climate change will be accounted for, ideally using the new projections due to be released in 2018.
O3 – Quantifying and contrasting the severity of events in selected urban centres with that for the UK as a whole, accounting for spatial correlation of assets at risk (i.e. the possibility of multiple sites getting flooded at once). This will be done through the construction of a set of probabilistic impact ‘footprints’, creation of an entry-level catastrophe model, and using a severity metric based on impact (e.g. financial cost).
Training and Skills
CENTA students will attend 45 days training throughout their PhD including a 10-day placement (e.g. at an insurer). In the first year, students will be trained as a single cohort on environmental science, research methods and core skills. Throughout the PhD, training will progress from core skills sets to master classes specific to the student's projects and themes.
Specifically for this project, the PhD student will gain state-of-the-art skills to analyse financial loss and meteorological data; such large datasets are sometimes referred to as ‘Big Data’. This training will be in modelling, planning for resilience, and understanding environmental systems: In particular in:
- Catastrophe models and their specific features to model hazards, vulnerabilities and exposure
- Fieldwork, integrated modelling,
- GIS, and relevant programming e.g., SQL, R, python.
- Extreme value and multivariate analytical statistics
The student will gain strength in skills identified by NERC as ‘most wanted’ for jobs in the environment sector; ‘modelling’, ‘multi-disciplinarity’, ‘numeracy’, ‘risk and uncertainty’.
This is excellent employment market preparation as scientific research skills, technical analysis and industry related model skills will be practiced and gained. Catastrophe modelling underpins all financial risk assessment due to natural hazards, and will soon become critical in Disaster Risk Reduction (DRR) and humanitarian efforts, ideally equipping the student to a range of careers upon completion of the project.
Year 1: Firstly, the student will develop an understanding of the sparse tide-gauge data and how to estimate severity between them using i) the SRJPM method ii) a storm surge model and iii) expectations from climatic data (i.e. ERA-40) [Befort et al., 2015]. Modelling expertise will be gained through (i) modelling a historical storm surge event using recorded data and an existing model (ADCIRC), tentatively the severe 2013 event in the Great Yarmouth area on the UK’s east coast; and (ii) statistical analysis of tidal gauging records held at the two stations in the area (Cromer and Lowestoft). Modelling and meteorological results will be combined with historical records at gauged locations to characterise storm surge extremeness for ‘synthetic gauges’ between gauging sites, creating an approach that allows rapid estimation of water height at the coast - all around the UK - for a large number of statistically representative events.
Year 2: In this year, the student will focus on modelling how water progresses inland using Dr Yu’s FloodMap software. Up to 1,000 flood footprints will be created at the UK scale, and finer-scale modelling using high-resolution LiDAR data at selected sites will allow some sources of uncertainty in the modelling to be considered. This builds on a current project modelling storm surge risks for New York.
Year 3: Finally, risk posed by the flooding will be assessed by developing an entry-level catastrophe model of coastal inundation. This framework uses the inundation footprints from Year 2, translating them into impacts (e.g. on critical infrastructures). In parallel, future impacts will be predicted by using climate projections to extend the modelling of Year 2: that is, understanding how the size and probability of the event footprints may change. We envisage including an industry-relevant resilience assessment through portfolio analysis (e.g. for railways).
Partners and collaboration (including CASE)
This project is a CASE studentship with the British Geological Survey (BGS), Keyworth, working with Prof Kate Royse (BGS Science Director – GeoAnalytics and Modelling). Within CENTA, the student will work closely between both Loughborough and Birmingham. Internationally, this work is supported by Prof. Ning Lin of the Hurricane Hazards and Risk Analysis Research Group at the prestigious Princeton University; it is envisaged that some of the training support in the CENTA student-support package might support direct collaboration with Prof. Ning, including a visit to Princeton University.
For information about this project, please contact Dr John Hillier (email@example.com). For enquiries about the application process, please contact the School of Social, Political and Geographical Sciences Research (firstname.lastname@example.org). Please quote CENTA when completing the application form: http://www.lboro.ac.uk/study/apply/research/.
Please contact Dr Dapeng Yu until January 2018 while Dr Hillier is on paternity leave