For the past 2000 years Earth's magnetic field has been weakening. At the going rate of decay, the north-south magnetic dipolegenerated within the convecting metallic fluid of Earth's outer corewould totally vanish, perhaps reversing polarity in the next 2000 years.
This scenario of a coming attempt by Earth's magnetic field to reverse its polarity is suggested by direct observation of the field since the 19th century and laboratory investigation of historic lavas and other fired materials that record the ambient field while cooling.
The ongoing weakening of the field does not ensure that a reversal will occur. After all, Earth's dipole reverses direction only on occasion, currently at a rate of a few times each million years. How a change in polarity is actually approached and, moreover, the degree to which such a process can be predicted, are unclear. Nonetheless, a significant step toward such an understanding may have been made through investigations of the ancient, or paleomagnetic field recorded in Ocean Drilling Project (ODP) marine sediment cores.
A 1993 paleomagnetic study of sediments obtained from ODP sites in the Pacific suggests that polarity reversal may be a more systematic process than previously thought. Researchers at the Institut de Physique du Globe de Paris found a rather remarkable pattern of intensity change (see Figure 1) during the past few million years, one in which the dipole field appears to re-energize immediately following a change of polarity. Following each revitalization, dipole field strength is observed to fluctuate while generally weakening at a more or less constant rate until polarity reversal again becomes likely. The figure suggests that this behavior occurred not only during successful reversals within the Matuyama reverse polarity chron when the field predominantly possessed normal (reverse) polarity, but also about 920,000 years ago when the field attempted to reverse but failed.
Fig. 1. Long-term asymmetric saw-toothed pattern: (above) Variations in relative paleointensity (expressed as VADMs, virtual axial dipole moments) across the normal-to-reverse (N -> R) Upper Jaramillo and reverse-to-normal (R -> N) Matuyama-Brunhes reversals obtained from ODP Leg 138 sediments from the equatorial Pacific. Open and solid circles are U-channel and discrete sample measurements, respectively (Courtesy of J.-P. Valet and L. Meynadier); (below) Schematic diagram of the general intensity pattern suggesting that field revitalization occurs not only during successful reversals, but also during unsuccessful, or aborted reversal attempts.
Such a process brings to mind an early statistical reversal model developed in the late 1960s. It proposed that field reversal could only occur when the dipole field can interact with the typically weaker, more complex nondipole part of the field. According to the model, this occurs when dipole strength, assumed to vary in a sinusoidal wavelike manner, is at an intensity low. The newly reported sediment data, however, suggest that any substantial strengthening of the dipole requires an attempt by the field to reverse and, hence, links field generation to the reversal process.
Earth is currently in the Brunhes normal polarity chron. By presenting globally distributed paleomagnetic intensity data from the Pacific, Atlantic, and Indian Oceans crossing the boundary of the last known reversal (the 780,000-year-old Matuyama-Brunhes reverse-to-normal polarity change), the investigators in Paris successfully tested their idea. These comparisons add credence to the claim that the asymmetric "saw-toothed" decay-and-recovery intensity pattern originally observed is a worldwide phenomenon linked to the dipole field. Partially compatible with this general picture are paleomagnetic data obtained from sequences of lava flows that recorded field behavior during reversals. These studies show that just following a polarity change the dipole field can strengthen to values well beyond that experienced prior to the flip. However, no long-term gradual field decay has been noted from lavas.
If the ocean sediment data are correct, then the process of polarity reversal of the geomagnetic field cannot be considered purely random, even though the time spanned between reversals itself appears to be random. Indeed, the researchers in Paris contend that the sediment data indicate a link between the time needed by the dipole field to weaken to the point of possible reversal and the degree of revitalization of the field following the previous reversal.
On another side of the issue is a statistical model developed in 1993. It is based on the known timing of field reversals over the past 100 million years and suggests that following a reversal there exists a 5,000 year-long dead-time during which there is zero probability of another occurrence. Then, according to the model, the probability of reversal steadily increases with time, but only for some 45,000 years. Although rooted in observed long-term field behavior, this model is not fully compatible with the French researchers' claim that the duration of a given polarity interval may be virtually determined by the initial field strength following reversal.
The sediment-derived finding of such a remarkable saw-toothed paleointensity pattern apparently linked to the reversal process has stimulated much discussion and debate among experimentalists and theorists who are exploring the source of Earth's magnetic field. Alternative nongeomagnetic explanations based on the process by which sediments become magnetized are also being sought. Whatever the outcome, this growing controversy promises to significantly affect our perceptions regarding field reversal. Whether we may one day understand the process sufficiently well to formulate predictions about the next polarity change is yet to be determined.
Return to Science and Society
Return to Starting Point