Equatorial dynamics of Earth's core
Principal investigator: Prof JD Fairhead
Sponsor: NERC
Value: £131449.30
Dates: 31st July 2005 to 30th June 2008
Summary
The Earth's magnetic field is generated by complex fluid motions in the Earth's liquid iron core. A major issue in geomagnetism for the last fifty years has been the question of whether wave motion is detectable in the core. In this context, wave motion would be responsible for the propagation (typically east-west motion) of magnetic features on the core surface whilst the underlying fluid has no net translation and hence no momentum transfer. Two recent studies (Jackson, 2003; Finlay & Jackson, 2003) have brought this to the fore, focusing attention on the equatorial belt which appears to be particularly unusual from an observational point of view. Coupled with theoretical arguments for the "equatorial waveguide" (Zhang, 1993) being a location for instabilities, the time is ripe for a thorough study of this region and its dynamics.
Obviously wave mechanisms are important in the atmosphere and oceans, and theory certainly suggests that similar mechanisms are possible in the core; yet crucial questions, such as even the sense of propagation of waves have to date only rudimentary answers (Hide, 1966; Jones et al, 2002). The identification of waves in the core could be vital in illuminating some of its physical properties, such as the unknown strength of the hidden toroidal magnetic field. This proposal represents a classic marriage of observation and theory to attempt to unravel modes of wave motion. It combines elements of inverse theory, spectral analysis, numerical simulation and statistical palaeomagnetic analysis to focus on one of the primary unresolved questions of geomagnetism.
Specifically we propose to make use of a recently-developed (Jackson, 2003), high-resolution imaging technique (Maximum Entropy imaging) to optimally exploit the large historical magnetic database at Leeds, in tandem with the tranch of new observations from the three satellites currently in orbit, Oersted, Champ and SAC-C. We will exploit these data by creating two homogeneous time-dependent models of the magnetic field at the core-mantle boundary. One model will be for the satellite era, 1965-2005, and the second for the historical period 1590-2005. Both have advantages and drawbacks: the shorter model has extremely homogeneous data coverage and very small errors, but is only for a short time span; the latter has more variable data coverage and errors, but obviously more temporal range.
We propose to analyse the resulting models to try to discern wave motion in the equatorial region. This will be carried out by frequency-wavenumber analysis, using techniques similar to those used in seismology. The primary issue is to attempt to detect dispersion, namely dependence of group velocity on wavenumber.
Complementary calculations will be carried out in the form of numerical instability analyses of rotating thermally-driven convection in the presence of a strong magnetic field, to determine the most unstable types of periodic instability. It is conjectured that modes exist which live in an equatorial waveguide, and observational evidence for this is presented in (Jackson, 2003), The dependence of these instabilities on several non-dimensional control parameters, as well as on the geometry of the imposed toroidal field, will be studied. The sensitivity of the eigenfunctions and eigenfrequencies to the strength of the toroidal field is expected to shed light on the ratio of poloidal to toroidal field strength in the core, still a relatively unknown quantity.
In a complementary study, we propose to statistically analyse the palaeomagnetic secular variation over the last few million years to try to detect latitudinal dependence indicative of different dynamical regimes.