Deep Earth Dynamics

The lower mantle plays a fundamental role in the thermal and chemical evolution of the planet. Much of the heat and heat producing elements ultimately driving continental drift and plate tectonics probably reside in the lower mantle. The boundary between the core and mantle is a primary interface within the deep interior and has a fundamental influence on the magnetic field, the cooling of the planet, and volcanism at the Earth's surface. Seismologists have revealed that the mantle side of this boundary is extraordinarily complex with km-scale fine structure, embedded with 10-km and 100-km scale layers of variable size and character. Thermal and chemical heterogeneity, solid-solid phase transitions, anisotropy, and melting within the lower mantle are probably all required in order to explain all these observed structures. Understanding the complexities of this region in a multi-disciplinary framework is essential to understand how the globally interconnected solid Earth system works. Computational geodynamics plays a fundamental contribution to understanding the deep mantle. We’ve been working with Don Helmberger, Jennifer Jackson and their students and post docs over the years as part of a long-running CSEDI – Cooperative Studies of the Earth’s Deep Interior – project. Here we describe some of our recent work in this area.

LLSVPs – is it a “High Bulk Modulus Layer”?

The largest discrete structures in the mantle are arguably the LLSVPs (the "Large Low Shear Velocity Provinces"), which seismic and mineral physics data suggest are thermo chemical. A number of geodynamic scenarios have been suggested (including passive piles, plume clusters, metastable superdomes, etc.).

Figure 1.Two-dimesnional models of compressible convection that illustrate the basic principle of a meta-stable super-plume. This concept was invented by former graduate student Eh Tan while he worked in our CSEDI project. For more details on these models, see Tan & Gurnis [2005].

One of the odd aspects of the LLSVPs is that there is an anti-correlation between shear and bulk sound velocity. But if the lay needs to be stable at the CMB, then the layer must have a high density and if the layer has a high density then it is not likely to be lifted off the CMB. It is possible to invent highly tuned sets of density and Rayleigh number so that the layer can lift off the CMB. But Eh Tan came up with a simple explanation that was consistent with the fundamental seismic observation that the shear and bulk sound velcoities were anti-correlated. The seismic results essentially demand that the Bulk modulus of the anomalous material is different (Larger) than ambient. This means that the material has a different compressibility.

Stabilized by an intrinsically larger density, a pile will remain at the CMB until exceeded by a thermal density difference of opposite sign. However, if there is a difference in compressibility between the plume material and ambient conditions, then upwellings become metastable (Figure 1). A sequence of models shows that if the zero pressure density is 2 to 3% larger and the adiabatic bulk modulus (Ks) is 4 to 8% larger than ambient mantle, then large metastable structures can form. The assumption of a larger Ks is not arbitrary: Given observed variations of δVS and δVP within the African LLVP, then Ks must be larger than the value in ambient mantle if the density anomaly (at the same temperature) is equal or greater than ambient conditions.

In the model shown in Figure 1, the anomalous material heats, becomes more buoyant than the background, and moves upward. During ascent, its buoyancy gradually decreases, due to an increasing adiabatic density difference, and rises to a height of neutral buoyancy. Above the HNB, the anomalous material becomes denser than the background and sinks. The structure stands high above the CMB, but remains metastable depending upon the equation of state and depth-dependence for the coefficient of thermal expansion

Spin-Transition in Iron Oxides

A high-spin to low-spin electronic transition of iron occurs in ferropericlase (Fp), a dominant lower mantle constituent. Mineral physicists, including Jennifer Jackson in the Seismo Lab have shown that this needs to a graduate change in the density of the material with pressure, in a sense like a gradual phase transition.
Figure 2. Temperature and densities that enhance the vigor of convection in the presence of the spin transition in Iron. See Bower et al. [2009] for details

Geophysics graduate student Dan Bower, now at ETH-Zurich, using a numerical model to explore the consequences of the intrinsic density change (2–4%) caused by the Fe2+ spin transition in ferropericlase on the style and vigor of mantle convection. The effective Clapeyron slope of the transition from high to low spin is strongly positive in pressure-temperature space and broadens with high temperature. This introduces a net spin-state driving density difference for both upwellings and downwellings. In 2-D cylindrical geometry spin-buoyancy dominantly enhances the positive thermal buoyancy of plumes. Although the additional buoyancy does not fundamentally alter large-scale dynamics, the Nusselt number increases by 5–10%, and vertical velocities by 10–40% in the lower mantle. Advective heat transport is more effective and temperatures in the core-mantle boundary region are reduced by up to 12%. Dan's findings are relevant to the stability of lowermost mantle structures.

Geography of the Deep Mantle

Figure. Four-dimensional dynamic earth model aimed at better understanding the geography of the LLVSP in the deep mantle from the work of geophysics graduate student Dan Bower
One way to realistically constrain the evolution of deep earth structures is through linking surface evolution to the CMB via 4-D dynamic Earth models (Figure 3). As an analogy, plate tectonics provides a context where fine scale geophysical, geochemical and geological observations are routinely interpreted within a large-scale context. Much of our understanding of the CMB rests with detailed seismic observations of discrete structures (e.g. D”, LVZ, ULVZ, LLSVSP, rolling hills), which can now be linked to the large-scale structure of the CMB, which we call the “seismic geography”, to that predicted by global models based on the history of subduction. Through the GPlates consortium, our group leads the effort to link our global models of mantle convection to modern paleogeographic tools. As part of his Ph.D. work, geophysics graduate student Dan Bower is exploiting this paradigm within a four dimensional system, called Dynamic Earth Models, which merge detailed high resolution paleogeographic plate tectonic reconstructions with time-dependent models of mantle convection.

Recommended Readings

Bower, D. J., J. K. Wicks, M. Gurnis, and J. M. Jackson, A geodynamic and mineral physics model of a solid-state ultralow-velocity zone, Earth and Planetary Science Letters, 303, 193-202, 2011.

Bower, D. J., M. Gurnis, J. M. Jackson, and W. Sturhahn, Enhanced convection and fast plumes in the lower mantle induced by the Fe2+ spin transition in ferropericlase, Geophysical Research Letters, 36, L10306, doi:10.1029/2009GL037706, 4 pp., 2009.

Tan, E., and M. Gurnis, Metastable superplumes and mantle compressibility, Geophysical Research Letters, 32, Number 20, L20307 10.1029/2005GL024190, 4 pp., 2005.

Ni, S., Tan, E., Gurnis, M., and Helmberger, D., Sharp sides to the African superplume, Science, 296, 1850-1852, 2002.

Convection Movies
Geodynamics Home | People | Projects | GPlates

Last Updated November 30, 2015
Copyright © 1997-2012 California Institute of Technology, Pasadena, California 91125, USA