R. Holme, ELECTROMAGNETIC CORE-MANTLE COUPLING - I - EXPLAINING DECADAL CHANGESIN THE LENGTH OF DAY, Geophysical journal international, 132(1), 1998, pp. 167-180
Measured changes in the Earth's length of day on a decadal timescale a
re usually attributed to the exchange of angular momentum between the
solid mantle and fluid core. One of several possible mechanisms for th
is exchange is electromagnetic coupling between the core and a weakly
conducting mantle. This mechanism is included in recent numerical mode
ls of the geodynamo. The 'advective torque', associated with the mantl
e toroidal field produced by flux rearrangement at the core-mantle bou
ndary (CMB), is likely to be an important part of the torque for match
ing variations in length of day. This can be calculated from a model o
f the fluid flow at the top of the outer core; however? results have g
enerally shown little correspondence between the observed and calculat
ed torques. There is a formal non-uniqueness in the determination of t
he flow from measurements of magnetic secular variation, and unfortuna
tely the part of the flow contributing to the torque is precisely that
which is not constrained by the data. Thus, the forward modelling app
roach is unlikely to be useful. Instead, we solve an inverse problem:
assuming that mantle conductivity is concentrated in a thin layer at t
he CMB (perhaps D ''), we seek flows that both explain the observed se
cular variation and generate the observed changes in length of day. We
obtain flows that satisfy both constraints and are also almost steady
and almost geostrophic, and therefore assert that electromagnetic cou
pling is capable of explaining the observed changes in length of day.