UKRI GCRF Tomorrow’s Cities hub project: Physics-based ground motion simulations for seismic risk assessment and uncertainty quantification in rapidly urbanising environments

Project background
You will participate in the UKRI Global Challenges Research Fund Hub “Tomorrow’s Cities” (https://www.tomorrowscities.org) targeting disaster risk reduction in rapidly urbanising environments, led by Prof John McCloskey. This £20M, 5-year research hub is founded on the principle that it is possible for the development of new urban environments to be undertaken in a risk sensitive manner. This PhD will contribute to the application of seismic wave simulation to inform risk sensitive development in Istanbul and Kathmandu. This is an exciting opportunity to be part of an ambitious research programme and global science partnership.

You will work closely with an Edinburgh based PDRA, Dr Karim Tarbali, and Prof Carmine Galasso at University College London in the application of wave simulation to quantify seismic hazard in rapidly urbanising regions around Istanbul and Kathmandu.

In particular, you will drill down on specific aspects of the process to better constrain and quantify key uncertainties as they emerge during the project. Given the high vulnerability of infrastructure within Nepal and Turkey, the aim is to provide high-fidelity simulation results that can be used in various engineering applications.

Specifically, you will have a strong focus in the effects of uncertainty in the seismic source on the resulting distribution of ground motion taking both a theoretical and simulation based approach. You will undertake an initial case study looking at the consequences of uncertainty in the seismic sources around Istanbul on the resulting estimate of ground motion. The results of this work can then be scaled to Kathmandu later in the project.

Research questions
How does uncertainty in the earthquake source model propagate through to uncertainty in the resulting ground motions, engineering demand parameters, and eventual risk measures?
Which sources of uncertainty are most important? These include the magnitude of the source; the kinematics of the slip distribution of larger events; the tradeoff between magnitude and distance from target regions; impacts of basin geometry and topography.
How can we quantify and communicate the key uncertainties?
These questions will be answered by exploring the scaling from artifical slip models where we can test specific hypotheses in a controlled environment through to the uncertainty in real world slip models.

More detailed site specific questions will be developed in conjunction with project partners in Turkey and Nepal as part of the project.

Methodology
Physics-based ground motion simulations propagate the energy released from an earthquake slip model, through a potentially complex model of the earth’s crust to estimate the ground shaking experiences in specific regions. The spectral element method is a computational intensive approach to modelling the time evolving wave-field in 3D. The potential geological complexity needs to be traded off against the challenges in parameterising such models against sparse data and the high computational cost of running complex models. For example, can we better understand uncertainty through running few higher resolution simulations or a larger ensemble of lower resolution models that explore a wider parameter space?

The aim of this project is to investigate the impact of 1) the aleatory variability in model parameters, and 2) the epistemic uncertainty due to the different modeling approaches on the variability in simulated ground motion properties (and consequent damage/loss metrics). You will work closely with a PDRA who is experienced in seismic wave simulation in order to learn the core skills and then apply these in new ways to quantify uncertainty.