Motions of the Earth's Core and Mantle, and Variations of the Main Geomagnetic Field

Abstract

Theoretical work on the magnetohydrodynamics of the earth's liquid core indicates (a) that horizontal variations in the properties of the core-mantle interface that would escape detection by modern seismological methods might nevertheless produce measurable geomagnetic effects; (b) that the rate of drift, relative to the earth's surface, of nonaxisymmetric features of the main geomagnetic field might be much faster than the average zonal speed of hydrodynamic motion of core material relative to the surrounding mantle; and (c) why magnetic astronomical bodies usually rotate. Among the consequences of (a) and (b) are the possibilities that (i) the shortest interval of time that can be resolved in paleomagnetic studies of the geocentric axial dipole component of the earth's magnetic field might be very much longer than the value often assumed by many paleomagnetic workers, (ii) reversals in sign of the geomagnetic dipole might be expected to show some degree of correlation with processes due to motions in the mantle (for example, tectonic activity, polar wandering), and (iii) variations in the length of the day that have hitherto been tentatively attributed to core motions may be due to some other cause.