- An opportunity to create compound (i.e. multi-hazard) event set containing very rare, yet plausible scenarios; 1 in 100 to 1 in 1000 years events for today’s climate.
- Examine extreme weather in future climates.
- Storyline approach to understand and explain illustrative scenarios of co-occurring hazard combinations, and potential for ‘real world’ impact (e.g. rail or insurance sectors).
How extreme might the severe risk posed by multiple weather-related hazards get? Worldwide, many countries are severely impacted by co-occurring or ‘compound’ hazards [Hillier, 2015; Zscheischler, 2018], but these questions are understudied, particularly when asking what the most severe multi-hazard scenarios might be. Illustratively, consider Australia and ENSO. During summer, El Niño tends to reduce the likelihood of tropical cyclones and flooding, but enhances drought and wildfire risk, whereas La Niña drives the opposite conditions. Under the present climate, there are then two regimes or ‘modes’ of hazard behaviour which seem unlikely to co-occur [Hillier, 2020].
- How severely might extreme wind and flooding combine, in a discrete event or across a season? And, why?
- How safe are we to assume that two ‘antithetic’ hazard modes offset each other, and might they sometimes exacerbate to amplify impacts?
- Are there rare, very severe, and as yet unrealized combinations of hazards could be devastating, such as tropical cyclone followed by deadly heat? [Matthews, 2020]
- How might these change in future climates?
So, a fundamental question that remains is the possible nature of these most severe hazard combinations. This project will robustly examine the plausible limits of co-occurring climate extremes using a recently developed method [McKinnon and Deser, 2018] that blends of statistical and dynamical climate modelling. To push beyond purely dynamical approaches (e.g. UNSEEN) [Thompson, 2017], it will simulate 10s of thousands of years by combining the observations or reanalysis products (e.g., CRU TS v4.05 or ERA5) and CMIP5/6 large ensemble simulations.
To build upon existing work [Hillier, 2015; Hillier, 2020], an initial focus will on wintertime flooding and extreme wind in the UK, Europe and then across the mid-latitudes globally. This would tie into existing work with the Bank of England and Sayers LLP, illustrating the opportunities for the application of this work. Exciting opportunities then exist to progress into other new areas of study by making the work highly multi-hazard (n > 2), continuing with the UK (i.e. adding extreme cold or snow) or selecting other study sites across the globe.
- Climate and Environmental Sustainability
Primarily, the project will employ the Observational Large Ensemble (Obs-LE) methodology [McKinnon and Deser, 2018] to produce a large (>1000-member) simulation ensemble by combining the observations or reanalysis products (e.g., CRU TS v4.05 or ERA5, 1950-2020) and the CMIP6 HighResMIP (Haarsma et al., 2016) high-resolution ensembles (up to 25 km). This generates a wider and denser spectrum of internal variability via bootstraping and spectral resampling of observed variability merged with the forced response of the CMIP simulations.
The method will be adapted for impact-based proxies (e.g. cumulative v3 over a threshold) [Hillier & Dixon, 2020] to make it most relevant for hazards, and validation will use the UNSEEN approach (e.g. SEAS5) as well as observations.
Once identified, explanations for patterns and characteristic events will be sought using methods most suitable to hazards and events e.g. Lagrangian tracking of air parcels.
Training and skills
The PhD student will gain state-of-the-art skills in the analysis of “big data”, including:
- Geographic Information Systems and computer programming (e.g. Python, R, Matlab)
- Environmental modelling concepts such as the critical interpretation of physically-based climate models, and appropriate statistical approaches.
They will also learn background to the key topics required to understand mid-latitude climate risks:
- Extra-tropical cyclones.
- Risk assessment and communication, including catastrophe modelling [Mitchell-Wallace et al., 2017], and the storyline approach.
Partners and collaboration
This project is a partnership with Dr Neven Fučkar of the Environmental Change Institute, University of Oxford, the Bank of England (high-level steer) and the Willis Research Network (WRN), a scientific network that works in (re)insurance to better understand and quantify natural hazard risks. Dr Neven is a climate scientist who is passionate about the dynamics, prediction, attribution and impacts of extreme events in a changing climate.
There will be opportunities to visit Oxford, engage with UK infrastructure providers (i.e. road, rail) through Prof. Chapman, and potential exists for a placement either at the WRN or the Bank of England.
For information about this project, please contact Dr John Hillier ([email protected]). For more information about CENTA and the application process, please visit the CENTA website: www.centa.ac.uk. For further enquiries about the application process, please contact the School of Social Sciences & Humanities ([email protected]). Please quote LU6_CENTA when completing the application form: http://www.lboro.ac.uk/study/apply/research/.
To create a secure, safe basis for the PhD, the student will start with a bivariate analysis of flooding and extreme wind in the UK, building upon and verifying results against a growing body of work [e.g. Hillier, 2020; Hillier & Dixon, 2020; Owen, 2021]. They will familiarize themselves through case studies (e.g. historical storms, stormy years), and use of historical data to, types of climate data to be used and by implementing the primary statistical technique. Calibration will be to loss data (e.g. rail) to create impact-based proxies [e.g. Hillier, 2020], fidelity tests will be used to verify that known behaviours (e.g. due to NAO) exhibit appropriately, and the UKCP18 climate projections will be useful for verification. The aim will be to identify events robustly.
The aims for the second year are two-fold. First, to complete the bivariate UK analysis, work will be undertaken to understand the atmospheric processes that strengthen (or weaken) the flood-wind correlation for the most extreme events. In parallel with this, the methodology will be made multi-variate (e.g. to also consider the ‘antithetic’ hazards of cold and snow) and/or extended to the mid-latitudes globally.
In the third year, the project will be guided by results, either selecting additional case study countries or pursuing the analysis to consider co-occurring extremes in future climates.
Haarsma, R. J., Roberts, M., Vidale, P. L., Senior, C. A., Bellucci, A., Bao, Q., Chang, P., Corti, S., Fuckar, N. S., et al., 2016, High Resolution Model Intercomparison Project (HighResMIP) for CMIP6, Geosci. Model Dev., 9, 4185-4208, DOI: 10.5194/gmd-9-4185-2016
Hillier, J.K., Macdonald, N., Leckebusch, G.C. and Stavrinides, A., (2015) ‘Interactions between apparently ‘primary’ weather-driven hazards and their cost’. Environmental Research Letters, 10(10). Doi: 10.1088/1748-9326/10/10/104003
Hillier, J.K. et al. (2020) ‘Multi-hazard dependencies can increase and decrease risk’, Nature Climate Change, 10, pp. 595–598. doi:10.1038/s41558-020-0832-y.
Hillier, J. K. and Dixon, R. S. (2020) Seasonal impact-based mapping of compound hazards Env. Res. Lett., 15, 114013 doi:10.1088/1748-9326/abbc3d
Matthews, T., Wilby, R., Murphy, C., (2019) An emerging tropical cyclone-deadly heat compound hazard, Nature Climate Change, 9, doi: 10.1038/s41558-019-0525-6.
McKinnon, K.A. and C. Deser, (2018) Internal variability and regional climate trends in an Observational Large Ensemble, Journal of Climate 31 (17), 6783-6802, DOI:10.1175/JCLI-D-17-0901.1
Mitchell-Wallace, K., Jones, M., Hillier, J. and Foote, M., (2017) Natural catastrophe risk management and modelling: A practitioner’s guide. John Wiley & Sons.
Owen, L. (2021) How well can a seasonal forecast system represent three hourly compound wind and precipitation extremes over Europe? Environmental Research Letters, 16, 074019.
Thompson, V. (2017) ‘High risk of unprecedented UK rainfall in the current climate’, Nature Communications, 8, p. 107. doi:10.1038/s41467-017-00275-3.
Zscheischler, J., Westra, S., Hurk, B.J.J.M. van den, Seneviratne, S.I., Ward, P.J., Pitman, A., AghaKouchak, A., Bresch, D.N., Leonard, M., Wahl, T., Zhang, X., (2018) Future climate risk from compound events. Nature Climate Change 8. Doi: 10.1038/s41558-018-0156-3
The project is primarily computer-based modelling, and can be conducted remotely if required. All of the supervisory team are now entirely familiar with hybrid working, and as such we do not envisage COVID-19 or similar being a barrier to the progress of this PhD.