On Backus' mantle filter theory and the 1969 geomagnetic impulse

Summary. We present a discussion of Backus’ mantle-filter theory in relation to the 1969 geomagnetic impulse (or jerk). Backus shows that mixing of harmonic modes may be responsible for the negative value of the squared mantle-filter smoothing time which is found with European observatory data, a feature that is not mathematically possible for a pure internal harmonic impulse. We show here, and illustrate with a simple numerical experiment, that this negative value may simply be due to a remaining external solar cycle effect. This, together with the observation that the 1969 impulse is both world-wide and synchronous, leads us to propose that: (1) the smoothing time and zero-frequency delay of the mantle filter, as defined by Backus, are both likely to be shorter than a few years; (2) in that case, the 1956 elbow observed in the length of day curve would not be related to the 1969 impulse; (3) electrical conductivity is likely to be lower than a few hundred ω-1 m-1 in more than 97 per cent of the volume of the mantle. We recall our earlier suggestion that the magnetic variations apparently lead the length of day curve by 10-15 yr, which can be accounted for in terms of a Bullard model of electromagnetic core-mantle coupling.