The cure for cults that want to deny others their freedom of speech is more freedom of speech -- Fredric Rice

Public message 3283 - SCIENCE Area 19:51 Wednesday 4-Sep-91
From: HENRY SHAW
To: MIKE DAVEY
Re: EARTH'S MAGNETIC FIELD 1

MB> I understand that 'magnetic mapping' (for lack of the real term)
MB> has produced evidence that the magnetic 'north pole' has shifted
MB> many times in the earths past leaving some forms of rocks with
MB> magnetic 'map' pointing to where the magnetic pole was at the
MB> time the rock was formed.

MB> Just out of curiousity, do you happen to know how many times this
MB> shift is evident and the average time interval between them, and
MB> when the last one was?

Many times. Dozens. But I don't know an exact number. The last major reorganization of the magnetic field was a reversal at about 700,000 years ago. There does not appear to be any regularity to the pattern of reversals. That is, they are not periodic. Recently, people have postulated that they are a "chaotic" (in the strict sense) phenomenon.

Our knowledge of the history of the earth's magnetic field, naturally enough, gets poorer as one goes back in time. This is due to several things: (1) older rocks are relatively rarer than younger rocks; (2) older rocks are more likely to have undergone more complex histories (e.g. undergone multiple heating events, which act to "reset" the remanent magnetism); (3) absolute errors (not percentage errors) on the age of samples increases with the age of the sample, making it difficult to do the correlations required for high resolution magnetostratigraphy.

The basis for paleomagnetic studies is the fact that many igneous rocks contain minerals that are naturally magnetic (magnetite, or "lodestone" is an obvious example of a common rock forming mineral like this, but there are others). These minerals are only magnetic below a certain temperature known as the "Curie point". Above this temperature, the minerals have no net magnetism. When a rock cools below this temperature, the minerals acquire a magnetic moment (i.e. become like little compass needles) aligned with the earth's field. This direction is "locked in" over a relatively small temperature range, and persists even if the the relative orientation of the rock and the earth's field change at some later time (either by movement of the field, or by movement of the rock).

[The following stuff is taken from a message I wrote to Pat Goltz in this echo several years ago.]

The magnetic field of the earth is not a static thing. It undergoes fluctuations on timescales from milliseconds to hundreds of millions of years. The high frequency fluctuations are due to external (to the earth) causes, mainly variations in the solar wind. The longer-term fluctuations are due to variations in the motion of the fluid outer core of the earth at a depth of ~2900km. (For a review of the subject, see: Courtillot, V. and LeMouel, J.-L., 1988, Geomagnetic Time Variations, Annual Review of Earth and Planetary Science, v16, pp389-476.)

It has been known since shortly after Gauss first measured the strength of the field in ~1830 that the magnetic dipole moment of the earth is decreasing. This decrease has averaged about 0.05%/yr since 1830. There are, however, significant fluctuations about this average and in about 1960, the rate of decrease increased significantly (see Fraser-Smith, A. C., 1987, Centered and eccentric geomagnetic dipoles and their poles, 1600-1985. Reviews of Geophysics, v25, pp.1-16.).

The magnetic field of the earth is *not* a simple dipole (the field associated with, say, an "ideal" bar magnet). Though the dipole field is the dominant one today at the surface, the whole field is a bumpy, lumpy sort of thing. Mathematically, one can describe the total field by a sum of functions known as spherical harmonics. [Conceptually, this is similar to describing the motions of a vibrating guitar string; there is the fundamen- tal frequency and there are also higher frequency "harmonic" vibrations.] As the dipole field of the earth decays, higher-order "bumps" in the field grow, which implies that the total energy stored in the field has remained approximately constant (Verosub, K., and Cox, A., 1971, Changes in the total magnetic energy external to the earth's core, Jour. Geomagnetism and Geoelectricity, v23, pp. 235-42). The field we see at the surface of the earth is probably the result of a the sum over the fields produced by a number of convective cells in the outer core. Each of which produces a poloidal field. As the size, shape, and location of each of these cells changes over time, the net field we see at the surface undergoes corre- sponding changes.

There is ample evidence that the present time-variation of the field (the decreasing dipole moment) has not always been the case. Over short timescales (thousands of years), one can determine the strength of the paleo-field by measuring the strength of the remanent magnetism of baked clay pots and bricks and comparing that to the intensity produced in the same samples by heating and cooling them in magnetic fields of varying strength in the laboratory. Because of the "bumpiness" of the field over the surface of the earth at any given time, one cannot infer the strength of the dipole from these types of measurements made at a single site; one needs data from around the world.

Such data have been obtained on artifacts dating back more than 12,000 years, and the resulting curve, consisting of 1167 individual determinations for the variations in the intensity of the dipole field shows that the dipole moment has decreased for the past 2000 years, extending the historical record. The decay over the past 1000 years has been particularly fast. From 6000 to 3000 years before present, however, there was a significant INCREASE in the strength of the dipole (McElhinny, M., and Senanayake, W. E., 1982, Variations in the geomagnetic dipole 1: the past 50,000 years. J. Geomag. Geoelect. v34, pp.39-51.).

Over longer periods of time (hundreds of thousands to millions of years), there is evidence that the dipole field dissapeared completely and then reappeared with the opposite polarity numerous times over geologic history. Such events are called magnetic reversals. Perhaps the most compelling evidence for the reversals comes from the ocean floor, where the record of reversals is is clearly laid out as a pattern of symmetric magnetic "stripes" flanking both sides of the mid-oceanic spreading centers. This is not meant to neglect the very large body of data from land that also contains the record of reversals.

Detailed studies of the period of transition from one polarity to another suggest that the transition takes 1-2 thousand years, once initiated. There is also evidence that prior to the reversals, there are larger-than-average fluctuations in the partitioning of energy between the various harmonic components of the field. There are also instances of similar fluctuations (called excursions) that are not associated with an actual reversal. These may be instances of "failed reversals", in which the convective motions of the outer core underwent a reorganization that did not result in a change in the net polarity of the field.

--- TBBS v2.1/NM
* Origin: Diablo Valley PCUG-BBS, Walnut Creek, CA 510/943-6238 (1:161/55)