Search for a project

Seismic detection of hidden primordial regions in the Earth's lowermost mantle

Dr Andy Nowacki (SEE), Dr Sebastian Rost (SEE), Dr Mike Thorne (University of Utah, USA)

Contact email:

The (proto-) Earth cooled from being in a completely molten state, to the current structure of a solid, convecting mantle overlying the liquid outer core.  Since then, convection processes driving plate tectonics have homogenised the mantle.  However, geochemical and geophysical data indicate that the mantle preserved pockets of unstirred material, and may even have remained molten for some time at the base of the mantle (Labrosse et al., 2007).  This is an attractive idea because geochemical observations imply that part of the mantle must be capable of storing some ‘primordial’ material which is only occasionally brought to the surface and has otherwise remained ‘hidden’ for billions of years (Hofmann, 1997).  Finding the source of the hidden material is perhaps the largest unsolved problem in deep Earth science: this PhD project will use novel seismic imaging techniques and simulations to detect and analyse candidate mantle structures. 

Global seismic tomography has revealed that two continent-scale regions at the base of the mantle—known as the ‘Large Low-Shear Velocity Provinces’ (LLSVPs; Figure 1)—have anomalously low velocities (Garnero & McNamara, 2008), which could be caused by their being hotter than and/or chemically different to the rest of the lowermost mantle.  The LLSVPs have been discussed as excellent candidates to be these ‘hidden reservoirs’ of primordial material, provided they do not dynamically interact with the rest of the mantle.  A key indicator of their being distinct from the mantle would be the presence of chemically-induced strong changes in seismic velocity at their edges; however tomography is insensitive to this because of the inherent smoothing in the models it produces.

This project will use seismic observations to test if the LLSVPs could be stable, chemically-distinct features by examining waves which traverse the lowermost mantle for signs of ‘multipathing’ (Figure 2).  This occurs when the velocity V varies strongly enough over a sufficiently short distance x (dV/dx ~ O(1) m/s/km), typically leading to multiple wave arrivals (when only one is expected in a smoothly-varying medium) which arrive at a different incidence angle or azimuth than usual (e.g., Ni et al., 2002).  Such arrivals are excellent probes of the location, boundaries and nature of structures in the deep Earth because the presence of strong velocity gradients strongly suggests that chemical variations are the cause, which indicates that LLSVPs may be separate from convection processes.

The student will make observations of the horizontal and vertical incidence angles of multipathed arrivals of seismic phases sensitive to velocity gradients in the mantle (such as Sdiff, Pdiff, SKS and SKKS) within, around and away from the LLSVPs, using dedicated seismic arrays (e.g., the Yellowknife, Canada and Warramunga, Australia arrays) and collections of broadband seismic stations which can be grouped together for this purpose (e.g., USArray, Hi-net, F-net).  By mapping the presence (and absence) of multipathing, we can create the first direct global picture of velocity gradients in the lowermost mantle, which can be compared to tomography and previous studies.  Advanced 3D modelling of waveform features will be used to quantify the shape and sharpness of these features, and determine better than before whether the LLSVPs could be as ancient as is suspected.

 Figure 1: (a) Seismic tomography of the lowermost mantle (Ritsema et al., 2010), showing the two Large Low-Shear Velocity Provinces (LLSVPs) as regions of low shear wave velocity (dVS).  (b) Horizontal gradient (dVS/dx) in velocity derived from (a).  These values are underestimates due to smoothing in tomography, but suggest that strong gradients may exist near the LLSVPs.


In this project, you will investigate the poorly-understood lowermost mantle using the study of seismic waveforms and array processing to determine how strongly heterogeneous the region is.  Depending on your interests and strengths, you may choose to focus in a number of areas to develop the work and make a novel contribution to the study of the deep Earth.

  1. Gather high-quality seismic datasets using seismic arrays and global broadband seismic networks, develop techniques to detect multipathing and measure discrepancies in incoming slowness and backazimuth between the data and 1D models (Rost & Thomas, 2009).
  2. Examine correlations between your map of regions of strong velocity gradients and global tomography, small-scale scattering studies (Rost & Earle, 2010), core–mantle boundary topography, and other observables.
  3. Use 3D, finite-frequency modelling to accurately determine the possible structures which are responsible for the seismic observations (e.g., Nowacki et al., 2016, and references therein).
  4. Apply, develop and extend an automated method to detect multipathed waves using stacks of vespagrams and cluster analysis to identify true seismic arrivals.
  5. Build a suite of models and the corresponding synthetic seismograms of the lowermost mantle against which seismic data can be compared automatically, again to seek out regions of high velocity gradients

Figure 2: Example of multipathing.  (a) Path of Sdiff waves recorded at the Kaapvaal array (green triangles) from an event in Fiji (orange circle), traversing the edge of the African LLSVP.  (b) Seismic data showing multipathing Sdiff arrivals when sorted by the azimuth from the event to the station.  (c) Frequency-wavenumber stack, with dark colours indicating strongest power, showing that energy arrives at the array with deviations in backazimuth and slowness compared to the theoretical arrival (black cross, labelled) for a 1D Earth model.

Potential for high impact outcome

This work is placed to push forward our understanding of the dynamics of the whole Earth system: determining the nature of the LLSVPs is a fundamental outstanding question for all of deep Earth science.  In addressing these questions you will be producing the highest-quality observations, models and hypotheses that will be published in the highest-impact journals.


The student will work under the supervision of Dr Andy Nowacki and Dr Sebastian Rost, within the Institute of Geophysics and Tectonics in the School of Earth and Environment, and with Dr Mike Thorne at the Department of Geology and Geophysics, University of Utah.  You will be an integral member of the Deep Earth research group in Leeds—one of the largest and most prestigious groups in the world examining how the Earth’s mantle and core behave—and have the opportunity to work with and learn from fellow researchers, both senior academics and fellow students, in a broad swathe of geophysics.  You will also have the chance to visit Utah during your PhD.

You will have the chance to develop expertise in analysing data and scientific programming, including using the largest supercomputers in the country, alongside developing the core transferable skills of project management, team-working and communication.  Opportunities to travel to international conferences will be part of the studentship alongside time spent in Utah.  You will have access to the broad spectrum of training courses provided by the Faculty that include an extensive range of workshops from numerical modelling, managing your degree, through to preparing for your viva (

Student profile

A good applicant should have a strong background in physics or quantitative Earth science and a passion for studying how the Earth works.  Interest and experience in seismology would be useful, as would some experience of programming, but not essential.


  • Garnero, E.J., McNamara, A.K., 2008. Structure and dynamics of Earth’s lower mantle. Science 320, 626–628. doi:10.1126/science.1148028

  • Hofmann, A., 1997. Mantle geochemistry: the message from oceanic volcanism. Nature 385, 219–229. doi:10.1038/385219a0

  • Labrosse, S., Hernlund, J., Coltice, N., 2007. A crystallizing dense magma ocean at the base of the Earth’s mantle. Nature 450, 866–869. doi:10.1038/nature06355

  • Ni, S., Tan, E., Gurnis, M., Helmberger, D., 2002. Sharp sides to the African superplume. Science 296, 1850–1852. doi:10.1126/science.1070698

  • Nowacki, A., Wookey, J., 2016. The limits of ray theory when measuring shear wave splitting in the lowermost mantle with ScS waves. Geophys J Int In press. doi:10.1093/gji/ggw358

  • Rost, S., Earle, P., 2010. Identifying regions of strong scattering at the core-mantle boundary from analysis of PKKP precursor energy. Earth Planet Sci Lett 297, 616–626. doi:10.1016/j.epsl.2010.07.014

  • Rost, S., Thomas, C., 2009. Improving seismic resolution through array processing techniques. Surv Geophys 30, 271–299. doi:10.1007/s10712-009-9070-6

Related undergraduate subjects:

  • Earth science
  • Geology
  • Geophysics
  • Geoscience
  • Mathematics
  • Physics