Jon E. Mound

Melting of the Earth's inner core.

The Earth’s magnetic field is generated by a dynamo in the liquid iron core, which convects in response to cooling of the overlying rocky mantle. The core freezes from the innermost surface outward, growing the solid inner core and releasing light elements that drive compositional convection. Mantle convection extracts heat from the core at a rate that has enormous lateral variations. Here we use geodynamo simulations to show that these variations are transferred to the inner-core boundary and can be large enough to cause heat to flow into the inner core. If this were to occur in the Earth, it would cause localized melting. Melting releases heavy liquid that could form the variable-composition layer suggested by an anomaly in seismic velocity in the 150 kilometres immediately above the inner-core boundary. This provides a very simple explanation of the existence of this layer, which otherwise requires additional assumptions such as locking of the inner core to the mantle, translation from its geopotential centre or convection with temperature equal to the solidus but with composition varying from the outer to the inner core. The predominantly narrow downwellings associated with freezing and broad upwellings associated with melting mean that the area of melting could be quite large despite the average dominance of freezing necessary to keep the dynamo going. Localized melting and freezing also provides a strong mechanism for creating seismic anomalies in the inner core itself, much stronger than the effects of variations in heat flow so far considered.

Constraints on core‐mantle electromagnetic coupling from torsional oscillation normal modes

[1] Decadal axial angular momentum variations in the Earth’s core are believed to be carried by the normal modes of torsional oscillations. Coupling with the mantle transfers angular momentum to the latter, leading to changes in length of day (LOD). Electromagnetic stresses at the core-mantle boundary (CMB) may be an important coupling mechanism as well as a source of dissipation for torsional oscillations. In this work, we investigate whether the observed spectra of fluid core velocities and LOD variations can be both explained in terms of the normal modes of torsional oscillations when the only coupling with the mantle is through electromagnetic stresses. We show that this explanation may be true when the magnetic field at the CMB is based on a downward continuation of surface observations, provided the conductance at the bottom of the mantle does not greatly exceed 10 8 S and small wavelength field features do not contribute more than approximately 25% of the total radial field at the CMB. A larger conductance or a higher amplitude radial magnetic field results in a damping of the normal modes of torsional oscillation that is sufficiently large that they should not be detectable. In particular, we show that this is the case for the conductance and radial magnetic field that are inferred from the Earth’s forced nutations. If these constraints are correct, the decadal periodicities in the fluid velocity and LOD must then represent the preferred frequencies of the excitation mechanism of torsional oscillations rather than the signature of the free modes. Citation: Dumberry, M., and J. E. Mound (2008), Constraints on core-mantle electromagnetic coupling from torsional oscillation normal modes, J. Geophys. Res., 113, B03102, doi:10.1029/2007JB005135.

Sea‐level change and true polar wander during the Late Cretaceous

We demonstrate that rapid changes in the Earths rotation vector can drive relative sea-level (RSL) variations in excess of 100 m in as little as 1 m.y. Stratigraphic constraints on sea-level change can thus be used as an independent test of true polar wander (TPW) events proposed on the basis of paleomagnetic evidence. Recent estimates of Late Cretaceous TPW include a period of rapid motion, however a corresponding period of large and rapid sea-level change is not consistently evident in our preliminary examination of the geological record. This suggests that the magnitude of Late Cretaceous TPW may have been overestimated, although a definitive conclusion requires a comprehensive study of high-resolution Late Cretaceous sea-level records. (Index Terms: 1239, 1527, 4556, 8159).