Abstract:
Recent paleomagnetic records suggest that the geomagnetic field intensity has a saw-tooth shape matching the succession of polarity intervals. An alternative hypothesis, that mechanisms linked to the acquisition of magnetization would induce similar saw-toothed records, has been tested by several simulations of post depositional reorientations of magnetic grains. Exponential functions used to date to describe post-depositional magnetization (pDRM) processes do not account for the combination of saw-toothed fluctuations and reasonable delays in the recording of the position of the reversals. At least half of the magnetization must be locked in within a few centimetres below the surface. If not, large delays, which are not observed in the data, are introduced in the stratigraphic positions of the reversals. In addition, the rest of the magnetization must be acquired over depths involving several tens of meters to duplicate a saw-toothed shape. These conclusions are reached with or without incorporating intensity variations across reversals. If the original signal is, in fact, asymmetrical then the pDRM must be very limited to remain coherent with the measurements, since the distortions and the offsets induced by the pDRM smoothing are considerably amplified. We conclude that simulations of saw-tooth patterns of relative paleointensity by pDRM processes have consequences that are difficult to reconcile with our present knowledge of the physical properties of deep-sea sediments. Above all, the hypothesis that the saw-tooth is an artifact of the magnetization acquisition process would have major implications for any sedimentary record of geomagnetic features.