The remanent magnetism of some recent varved sediments

Abstract

Some unconsolidated water-laid sediments acquire a remanent magnetization as a result of the orientation during settling of the small fraction of their particles that have a permanent magnetic moment. Their alinement is affected by factors other than the direction of the geomagnetic field, such as the slope of the bed on which they are deposited, and the velocity with which the water moves over it. The paper describes measurements of the remanence of Swedish and Icelandic varved sediments, with average particle sizes lying between 5 and 25 μ , in which these factors were studied. Measurements of the remanence of these sediments redeposited in the laboratory under controlled conditions were also made. In these artificially deposited sediments the inclination of the remanence may be some 20° less than that of the magnetic field, but this ‘inclination error’ seems to be never more than 5 to 10° in natural sediments. In neither case was the dependence of inclination error on particle size suggested by earlier work confirmed. Under both natural and artificial conditions deposition on a sloping bed produces a deviation of the remanence from the direction for deposition on a horizontal surface. The direction of rotation of the remanence is the same as that produced by tilting a bed subsequent to deposition. The deviation is less under natural than under artificial conditions, never exceeding about twice the angle of dip of the bed. Experiments on deposition in running water indicate that velocities up to about 30 cm/s produces deviations of the remanence of up to about 20° in a direction related to that of the current flow, the magnitude of the effect being almost independent of velocity over the range of velocities between 5 and 30 cm/s. Size and shape analyses of the sediments show that the magnetic particles have a size distribution very similar to that of the sediment as a whole, and that they are not clearly divided into well-contrasted groups of spherical and plate-like particles, as was suggested in an earlier paper. A theory is proposed to account for the known facts of depositional magnetization. Deviations of the remanence from the field direction are explained mainly as a result of the rolling of particles on the bed during the last stage of settling. To a limited extent quantitative agreement between theory and observation is attained, but the process is clearly too complex to be adequately described in simple geometrical terms. It has not therefore been found possible to use the theory for the purpose of prediction. From a review of the evidence it would seem to be possible in principle to determine the magnetic secular variation from varved sediments, but a very large number of experiments would be needed for useful results to be obtained.