You will carry out laboratory experiments to quantify the quality of forecasts for catastrophic failure in rock samples in real time using wave-based techniques. You will use automated and state-of-the-art techniques of real-time data assimilation to compare the performance of wave-based models with discrete seismic event-based alternatives.
Seismic waves have been used to monitor and detect the onset of catastrophic events (e.g. earthquakes, landslides or volcanic eruptions). However, all of these results have been obtained in retrospect, i.e. given knowledge that the event already occurred: researchers looked back at the recorded data retrospectively, to find signals that reflected the event’s onset or imminence. This is a fundamental bias in the scientific process, and to be rectified, methods and models must be tested objectively (without interference from humans) on events that may or may not occur in the real-future.
In this project, you will develop methods to perform real-time wave- and event-based forecasts of material failure in the Real-Future Prediction & Control (RFPC) laboratory being built in Edinburgh. The RFPC consists of (i) an on-line repository for models, (ii) a laboratory in which catastrophic events may or may not be generated through applied stress on material samples (solids, and rocks in particular), and from which real-time data stream out, and (iii) a high-performance computing facility (the Terracorrelator) which performs analysis of the streamed data in real-time.
You will run a series of rock deformation experiments to test and develop models that predict bulk medium properties and fracturing events, and changes in their properties over time. Successful models will then be tested on real-future data from the field, using seismological data continually streaming into the Terracorrelator from around the world (selected as appropriate to the model) to predict future fracturing events in the real Earth.
This PhD has a generous EU-level salary, funds for international research visits & conferences, and is part of a European network of 15 PhD students working on related topics and with whom you will regularly interact.
A comprehensive training programme will be provided, comprising both specialist training and generic transferable and professional skills. You will be trained in experimental laboratory rock physics, high-performance computing, and automated data assimilation. You will also acquire skills that can be applied to a range of natural and induced hazards, as well as to other applications of real-time data assimilation.
Apply on-line at: http://www.ed.ac.uk/schools-departments/geosciences/postgraduate/phd/programmes-supervisors/physical-sciences/phd-projects?DisciplineID=3#projectlist
For more information, please contact Dr Andrew Bell ([email protected]