The late 1960s secular variation impulse, the eleven year magnetic variation and the electrical conductivity of the deep mantle

Annual mean values of geomagnetic field components from world observatories for the period 1947–77 are analysed in the hope of being able to separate true internal secular variation from remaining external effects. The simple representation of internal secular variation which is used (the fitting of low degree polynomials) is that used by Courtillot & Le Mouël (1976a) and advocated by Vestine and colleagues as early as 1947. The case of European observatories is treated in detail: the density of these observatories provides an excellent test for the method. Two major results come out of this study: (1) A large secular variation impulse is shown to have occurred over much of the northern hemisphere in the late 1960s. A very rough description of the impulse field is a supplementary westward rotation of parts of the outer core layers of approximately 0.1° yr−1. The impulse occurs in a very short time, certainly less than 4 yr. (2) The clean representation of internal secular variation allows in turn isolation of the solar-cycle related variation. The corresponding field is a meridional one; measurement of the amplitudes of the vertical and north components yields an estimate of the response of the conducting Earth to such long period signals. The modulus and phase of the response (i/e) are respectively 0.17 and 44°. The observation of the secular variation impulse provides constraints on deep mantle conductivity ‘from the bottom’, while the 11-yr response provides constraints ‘from the top’. It is found that conductivity does not exceed 100 Ω−1 m−1 in the mantle. A survey of possible conduction mechanisms and estimated values of associated parameters in the deep mantle suggest temperatures around 3500 K. However, physical parameters for the electronic semiconduction process are so uncertain that we may conversely use conductivity and temperature estimates to constrain, for example, the activation energy of lower mantle materials.