Skip to content
Dr Matthew Price

Dr Matthew Price

Research Associate

School of Earth and Ocean Sciences

+44 (0)29 2087 4336
Room 2.23C, Main Building, Park Place, Cardiff, CF10 3AT

From just a few kilometres below our feet, to 3000 km beneath us lies Earth's mantle. The mantle amounts to over 80% of Earth's volume, and near 70% of its mass, linking the deep heart of our planet, the core, and the crust at the surface. It plays a key role in many processes such as plate tectonics, volcanic hotspots, the transfer of heat from the interior to the surface, and as storage for a variety of different elements and compounds.

The structure of Earth has changed greatly over the 4 billion years since its creation, yet we only have a detailed view of the mantle structure at present day. My research attempts to further our knowledge of the role of the mantle throughout Earth's history. I do this using 3-D spherical models of Earth's mantle, coupled with inputs from a variety of other branches of Earth Science, to help us understand the evolving structure of the mantle through time and our present-day observations. By gaining a better understanding of these processes here on Earth, we are also able to better understand the limited observations of our planetary neighbours such as Venus and Mars and unravel their own unique geological history.


  • Mantle processes
  • Geodynamics
  • Numerical modelling
  • Fluid dynamics
  • Data assimilation
  • Inverse theory


PhD – School of Earth and Ocean Sciences, Cardiff University (2016)

BSc – School of Mathematics, Cardiff University (2012)

Career Overview

Postdoctoral Research Assistant – School of Earth and Ocean Sciences, Cardiff University (2016 - present)






I am currently a Post-Doctoral Research Associate working as part of the “Mantle volatiles: processes, reservoirs and fluxes” consortium led by University of Oxford, part of the 5 year Natural Environment research Council (NERC) – “Volatiles, Geodynamics and Solid Earth Controls on the Habitable Planet” Programme.

The Deep water cycle

The focus of my current role is to investigate the deep-Earth water cycle using 3-D numerical models. By adapting our mantle convection code TERRA to account for the influences of water within the mantle (on processes such as density and viscosity variations as well as melting temperatures), we can test and constrain the very important but currently very poorly understood deep water cycle (e.g. Price, Davies & Panton, 2019). 

Deciphering the deep mantle structure

Seismic imaging of the mantle reveals two significant regions of anomalously slow seismic waves (known as the large low shear velocity provinces – or LLSVPs) on the core-mantle boundary. The nature of these LLSVPs and whether they are purely thermal or thermo-chemical, is an open question in the earth science community. These LLSVPs have also been credited as the source of distinct geochemical signatures in hotspot lavas located at the surface above them, leading to the idea that the LLSVPs could also be reservoirs for primordial material.

By using 3-D mantle models, together with inputs from a range of geoscience (including information from seismic, geochemical and mineral physics data) I am interested in attempting to better understand the makeup and long term evolution of the deep mantle flow and it’s influence on these LLSVPs and its role in observed geochemical heterogeneity (e.g. Barry, Price et al., 2017).

The initial condition problem in mantle convection modelling

I am also interested in the problem of time dependence in mantle modelling, with my PhD focusing on understanding the time-dependent nature of this convection and the importance of initial conditions choice.

One way to better constrain the initial condition is through an inversion of the equations governing mantle convection. I have incorporated such inversion methods, which account for the wealth of present day information of the mantle (e.g.  seismic structure, geoid data), into a mantle model in order to constrain the initial condition. It has been shown that for one such inversion method, known as the adjoint, that there is an upper limit on the time interval which such solutions will practically converge (Price & Davies, 2018).