Uncertainty characterisation of ground motion from large earthquakes in urban areas

Large earthquakes are rare events. Even in regions of high seismicity, like California or Japan, the frequency of experiencing such earthquakes is typically expressed in a generation (i.e. 30 years). Because of their cost, seismic sensor networks are rather sparse, with stations spacing of 20-30 km at best. This inevitably means that when an earthquake does occur somewhere in the world, it is poorly recorded, particularly near to the epicentre. In urban areas, low-cost sensor networks, such as those in consumer grade instrumentation, offer one solution to provide more dense recordings of earthquakes – however, their use comes with severe limitations of accuracy and precision.

Statistical and numerical algorithms that can rigorously process the data collected from low-accuracy/precision devices are urgently needed. Epistemic uncertainty arises from the inability to gather sufficiently precise/accurate measurements. These algorithms will allow for coherent seismic characterisation from such devices, as all the uncertainties will be carried through the computations. Error propagation strategies will be developed to rigorously characterise the ground motion from different devices with given specifications. This will encourage scientists to have an “inclusive” approach to seismic data collection, in opposition to discarding potentially useful data as a result of misclassification. The analysis of the data on a network scale will give scientists and engineers a better insight into what is the cause/effect of strong shaking in different urban areas. The key parameters that define and limit the extreme ground motion will be investigated. The ground motion obtained from such characterisation will then be converted into input stochastic excitation and propagated through engineering models to assess the fragility of urban critical installations.

The PhD candidate will develop open-source numerical libraries for the unified modelling of epistemic and aleatory uncertainty in ground motion time history data. This will be linked to existing geotechnical and geophysical models of the near-surface and crustal structure for the assessment of the expected ground shaking in the near field. The student will join a multi-disciplinary research group at the Institute for Risk and Uncertainty, and will take active part in the activities running within the URBASIS-EU project.

The student will be based at the Institute for Risk and Uncertainty and the EPSRC ESRC Centre for Doctoral Training for the Quantification of Risk and Uncertainty in Complex Systems and Environments at the core of the University of Liverpool campus.

https://urbasis-eu.osug.fr/

https://riskinstitute.uk/researchprojects/urbasis

Requirements
Good programming skills in scientific computing with Python or Java. Understanding of practical C programming is also desirable. Knowledge of concepts in seismology and engineering seismology and Python programming is an advantage. Applicants should have a degree in Engineering/Physics/Geophysics/Mathematics. Excellent undergraduate and Master's degree grades are expected as well as good written and spoken English.

Funding details
The position is funded by the European Commission (EC) with salaries according to EC and local standards, minimum gross wage is 3500 euros before local taxes.

Starting Date
May 1, 2019

Student nationality
Please refer to the fee structure of the University of Liverpool.

Application
Candidates are requested to submit a single pdf file containing:

● A letter motivating the application

● A detailed CV

● Academic transcripts

to: Dr Marco De Angelis (marco.de-angelis@liverpool.ac.uk). 

 

Application deadline:  23 February, 2019, 10 p.m. Europe/London

Centre for Doctoral Training in Quantification and Management of Risk & Uncertainty in Complex Systems & Environments is funded by EPSRC and ESRC.