Seismic investigation of the Earth’s mantle with fibre-optic sensing

The Earth’s deep interior is a dynamic place, playing host to subducting slabs sinking to the core–mantle boundary ~3,000 km below the surface, mantle plumes rising from the deep, and the interaction of these modes of convection over millions of years, eventually causing plate tectonics and volcanism.  Although the broad-scale features of mantle dynamics are known, there remains much uncertainty over the details of convection (Garnero & McNamara, 2008) and in many ways we know more about the interior of the Sun than our own planet—because light travels through the outer layers of stars.  In this project, you will help to address the uncertainties in Earth structure by using a new type of seismic observation—recordings from fibre-optic arrays.  In a way, you will use light to see into the solid Earth.

The use of optical fibres to measure the strain along the cable and interpret this in terms of seismic waves is known as distributed acoustic sensing (DAS), and the method has found wide application in the monitoring of subsurface processes on the local and regional scales (e.g., Booth et al., 2022).  However, despite its potential (Wuestefeld et al., 2023), thus far little use of DAS data has been made for problems in global seismology.  This has partly been due to the difficult, large data volumes produced by high-frequency, long-length DAS recordings, and partly because of the higher noise inherent in fibre-optic data.  However, if DAS can be used for global seismic monitoring, it has the potential to vastly increase the amount of data which can be used for the study of the Earth because of the thousands of km of existing fibre-optic cable which could be used, including sub-sea telecommunications cables (Sladen et al., 2019).  Its application is also not limited to understanding Earth structure—the use of seismology to detect and characterise earthquakes and man-made explosions is also crucial for many reasons, including to monitor adherence to the Comprehensive Nuclear-Test-Ban Treaty (CTBT; Bowers & Selby, 2009).  This project therefore is a collaboration with AWE Blacknest, the organisation tasked in the UK with ensuring that nuclear tests which breach the Treaty are detected and understood.

As part of the Global DAS Month (Wuestefeld et al., 2023), we in Leeds and AWE Blacknest collected 12 TB of continuous DAS data along 40 km of fibre installed at the Eskdalemuir seismic observatory in Scotland, using our own DAS equipment (Figure 2a).  We recorded the M7.8 6 February 2023 Kahramanmaraş earthquakes in Turkey and Syria (Figure 2b–c), as well as other teleseismic events.  These data provide an ideal testing ground to investigate teleseismic DAS data, which can be compared to the data recorded by the conventional seismometer array, plus our local array of seismic nodes placed adjacent to the fibre-optic cable.  We also have access to the continuous recordings from other DAS arrays around the world, making comparisons between DAS experiments possible in terms of their geometry and acquisition parameters.

In detecting teleseismic earthquake arrivals, you will develop new skills and develop new methods.  The use of machine learning (ML) will be important in this work, for example to denoise the data (e.g., Lapins et al., 2024) or process data automatically (e.g., Ward et al., 2021).  The extraction of normal modes (whole-Earth oscillations) may also be possible.

Project Structure

In this project, you will work with leading researchers at Leeds with backgrounds in seismology and global geophysics, and expertise in DAS data.  The exact scope of the project will be developed with your own personal research interests in mind, but an initial plan could be:

1. Use the University of Leeds DAS data recorded at Eskdalemuir to measure P-, S- and surface waves from the 6 February 2023 Kahramanmaraş earthquakes, finding optimal processing parameters (e.g., filters, both in time and in space) or machine learning methods (e.g., Lapins et al., 2024) which make the data clearest.
2. Use array processing methods (Rost & Thomas, 2002) to identify the precise direction of travel of these waves and classify arrivals.
3. Compare the propagation angles of the measured arrivals with predictions from global models of the Earth (similarly to Ward et al., 2020).
4. Extend observations to other, smaller earthquakes which occurred during February 2023.
5. Alternatively, look for normal mode observations, or focus on event characterisation.
6. A further alternative would be to characterise the ground beneath the array using the DAS data, EKA array, and seismic nodes deployed as part of the Global DAS Month.

In addition to the above, we would support you in identifying any sites where existing fibre-optic cables could be used for a new or repeat DAS deployment, potentially gathering a wholly-new dataset for the project.

Impact of this work

The structure and dynamics of the deep Earth remain in crucial aspects mysterious, and if we are able to use DAS arrays to improve our understanding of these topics, there is great potential to solve these mysteries.  In addition, observatories around the world such as that at Eskdalemuir are needed to monitor natural earthquakes and human-caused seismicity.  In particular, the monitoring of adherence to the CTBT which bans nuclear tests relies on seismic arrays; improving our ability to use DAS data to complement these could vastly improve how closely CTBT adherence can be monitored.  Finally, DAS is a technology which has exploded in popularity over the last decade, and work in this area attracts much attention.  For these reasons it is expected that successful candidates will be in a strong position to publish their work in high-impact journals in a number of articles.  Global interest in the work will also allow for travel to international conferences to present your findings.

Applicant suitability

This project would suit candidates with an interest in the fundamental way in which the Earth behaves, and a drive for interrogating datasets by both using existing computational seismic techniques and possibly developing new ones.  You will need to be able to collate seismic data, process it using available tools, develop new tools where appropriate, and model your data using existing modelling workflows.  Candidates will usually be armed with undergraduate training in areas such as geophysics, physics, geology, applied mathematics and similar branches of quantitative science.  Programming experience is advantageous.

Training environment

You will be a key member of a team of researchers across the School of Earth and Environment tackling fundamental problems in the study of the solid Earth.  You will be part of the Deep Earth and Seismology research groups in the Institute of Geophysics and Tectonics, interacting daily with not only your supervisors, but other senior colleagues, postdoctoral researchers and fellow PhD students.  You will also have the opportunity to devise and co-supervise research projects for undergraduate and masters students.  In this project, you will be trained in many transferrable scientific skills, including analysis of large datasets, high-performance computer modelling, probabilistic inversion of geophysical data, and the dynamic communication of your ideas.  Study for a PhD in the Deep Earth and Seismic groups involves international travel to conferences and the possibility of seismic fieldwork.  In this project you may have the opportunity to work at AWE Blacknest for an extended internship.  You will be supported in identifying a field site and obtaining the resources to deploy the group’s fibre-optic sensing equipment and fleet of seismic nodes to gather new data, if appropriate.  As a member of the YES Doctoral Training Network, you will receive tailored training alongside a cohort of other postgraduate researchers.

References and further reading

• Booth, A.D., Christoffersen, P., Pretorius, A., Chapman, J., Hubbard, B., Smith, E.C., de Ridder, S., Nowacki, A., Lipovsky, B.P., Denolle, M., 2022. Characterising sediment thickness beneath a Greenlandic outlet glacier using distributed acoustic sensing: preliminary observations and progress towards an efficient machine learning approach. Annals of Glaciology 63, 79–82. https://doi.org/10.1017/aog.2023.15
• Bowers, D., Selby, N.D., 2009. Forensic seismology and the Comprehensive Nuclear-Test-Ban Treaty. Annual Review of Earth and Planetary Sciences 37, 209–236. https://doi.org/10.1146/annurev.earth.36.031207.124143
• Garnero, E.J., McNamara, A.K., 2008. Structure and dynamics of Earth’s lower mantle. Science 320, 626–628. https://doi.org/10.1126/science.1148028
• Lapins, S., Butcher, A., Kendall, J.-M., Hudson, T.S., Stork, A.L., Werner, M.J., Gunning, J., Brisbourne, A.M., 2024. DAS-N2N: machine learning distributed acoustic sensing (DAS) signal denoising without clean data. Geophysical Journal International 236, 1026–1041. https://doi.org/10.1093/gji/ggad460
• Rost, S., Thomas, C., 2002. Array seismology: Methods and applications. Geophys. 40, 1008. https://doi.org/10.1029/2000RG000100
• Sladen, A., Rivet, D., Ampuero, J.P., De Barros, L., Hello, Y., Calbris, G., Lamare, P., 2019. Distributed sensing of earthquakes and ocean-solid Earth interactions on seafloor telecom cables. Nat Commun 10, 5777. https://doi.org/10.1038/s41467-019-13793-z
• Ward, J., Nowacki, A., Rost, S., 2020. Lateral velocity gradients in the African lower mantle inferred from slowness‐space observations of multipathing. Geochemistry, Geophysics, Geosystems 21, e2020GC009025. https://doi.org/10.1029/2020GC009025
• Ward, J., Thorne, M., Nowacki, A., Rost, S., 2021. Automatic slowness vector measurements of seismic arrivals with uncertainty estimates using bootstrap sampling, array methods and unsupervised learning. Geophysical Journal International 226, 1847–1857. https://doi.org/10.1093/gji/ggab196
• Wuestefeld, A., Spica, Z.J., Aderhold, K., (and other authors including Nowacki, A.), 2023. The Global DAS Month of February 2023. Seismological Research Letters 95, 1569–1577. https://doi.org/10.1785/0220230180