The aim of this project is to combine geophysical measurements with methods of Artificial Intelligence to monitor and forecast of volcanic eruptions.
Approximately 10% of the Earth’s population lives under the threat of one of about 1500 active volcanoes. Understanding the nature and impact of volcanic hazards, development of systems able to detect and track eruptive activity in real-time, and the implementation of frameworks to forecast future eruptions are key tasks for effective early warning and successful risk mitigation. At most volcanoes, geophysical unrest may provide evidence of forthcoming eruptions. Seismology, in particular, is unique amongst the Earth Science disciplines involved in volcano studies as it provides real-time information; as such, it is the backbone of every monitoring program worldwide. On the other hand, our understanding of the evolution of volcanic unrest and ability to forecast eruptions is hindered by the lack of a unified, quantitative, framework for classification and interpretation of volcano-seismic signals. Utilization of the unique attributes of seismology and integration within multidisciplinary dataset, models and frameworks for the investigation of volcanic processes is central to this project; the activities proposed here will contribute to developing statistically meaningful models of the evolution of volcano-seismic unrest during eruptive crises.
The candidate will work on developing a framework for understanding the seismic fingerprint of magma transport and eruption at active volcanoes and to evaluate the inherent forecasting potential of seismic and geophysical time series during volcanic crises. The project will address three major research questions central to hazard assessment and risk mitigation at active volcanoes:
1) Can current volcano-seismic classification schemes be extended or adapted to be universal?
2) Are the characteristics of unrest that does not culminate in eruption similar to those of activity that portends eruption?
Can we separate them, and if so, how?
3) Can we assimilate processes and signals at different volcanoes, or should individual systems be treated as unique?
The answers to these fundamental questions lie at the heart of every effort to decipher and fully exploit the forecasting potential of volcano-monitoring data. In this project the student will take an innovative approach to the analysis of volcanic unrest by adopting well-established techniques borrowed from fields such as image and speech processing, medical imaging, finance, robotics, and data analytics. They will apply methods of Machine Learning (ML), a branch of Artificial Intelligence based on the idea that computers can learn from data, identify patterns, and make decisions with minimal human intervention. The output of ML processing will be used to test statistical methods for eruption forecasts.
To apply for this opportunity please visit: https://www.liverpool.ac.uk/study/postgraduate-research/how-to-apply/
and click the ‘Apply online’ button.
Full funding (fees, stipend, research support budget) is provided by the University of Liverpool for 3.5 years for UK or EU citizens. Formal training is offered through partnership between the Universities of Liverpool and Manchester. Our training programme will provide all PhD students with an opportunity to collaborate with an academic or non-academic partner and participate in placements.
Salvage R.O., Karl S., Neuberg J.W. (2017) Volcano Seismology: Detecting Unrest in Wiggly Lines. In: Gottsmann J., Neuberg J., Scheu B. (eds) Volcanic Unrest. Advances in Volcanology. Springer, Cham
McNutt, S. R., Thompson, G., Johnson, J., De Angelis, S., Fee, D. (2015). Seismic and Infrasonic Monitoring. In The Encyclopedia of Volcanoes (pp. 1071-1099). Elsevier. doi:10.1016/B978-0-12-385938-9.00063-8McNutt, S.R., Volcano Seismology, Annual Review of Earth and Planetary Sciences, 2005 33:1, 461-491.
Bueno, A., Benitez, C., Diaz-Moreno, A., De Angelis, S., Ibanez, J. M. (2019). Volcano-Seismic Transfer Learning and Uncertainty Quantification with Bayesian Neural Networks, IEEE Transactions on Geosciences and Remote Sensing (in press)
Bueno, A., Diaz-Moreno, A., De Angelis, S., Benitez, C., Ibanez, J. M. (2019). Recursive Entropy Method of Segmentation for Seismic Signals. Seismological research letters, 90(4), 1670-1677. doi:10.1785/0220180317
Woollam, J., Rietbrock, A., Bueno, A., De Angelis, S. (2019). Convolutional Neural Network for Seismic Phase Classification, Performance Demonstration over a Local Seismic Network. Seismological Research Letters. doi:10.1785/0220180312