The aim of this paper is to give a brief description of the Earth’s magnetic field and the models associated with it. I will be looking at two models; one which is held by the vast majority and the other which I feel is more reasonable.
From very early times the effects of the Earth’s magnetic field have been observed, particularly in the use of compasses and in the spectacular auroral displays near the poles. However, it has only been in more recent times that the laws which govern magnetism and also the strength of this field have been evaluated. At the surface of the Earth this field is very weak compared to many fields which man can produce. However, because of the large distance and the large permeability involved, the quantity known as the magnetic moment is much greater than anything man has produced. B (the field strength) is of the order of 3 x 10-5 Tesla while M (the magnetic moment) is of the order of 8 x 1022 amp. metre.1
The magnetic field has a direction with three components (one North-South, one East-West and one towards the centre of the Earth) and a strength. It has been observed that mappings of lines with equal horizontal intensities show no resemblance to the broad features of geography. On the other hand, it has been shown that this field at a location varies its direction and its strength with time. No general pattern for this change has been found as the changes vary from place to place and from one time to another. The overall pattern of the Earth’s field indicates that it is approximately a uniformly magnetized sphere. Due to local variations, though, the overall picture is one of a field produced by an internal dipole with interference to the regular pattern being produced by local magnetization in the crust.
Causes of the field
There are three possible causes of this magnetic field. The first is that the Earth is a permanent magnet. This is quickly dismissed for two reasons:
- Permanent magnetism is lost when a substance is heated above a temperature known as the Curie point, and all but twenty-five miles of the crust is believed to be above the Curie point.
- The measured magnetism of the crustal rocks indicates that they do not supply nearly enough field strength to account for what we have.
The second possibility is that the rotation of a gravitational body may generate a magnetic field. This implies the discovery of hitherto unknown laws, these laws being effective only on a stellar scale. More knowledge on star behaviour has caused the originators of this theory to abandon it and thus, this also is ruled out.
Electric currents produce this field
The third and usually accepted notion is that electric currents in the core of the Earth produce this field. What is known about the nature of the Earth's core suggests that it would provide a very good environment for electric currents. That is, that the Earth's core consists of molten iron which has only small resistance even though resistance increases with temperature. The only question that remains is: what causes these currents? This is where the two models branch off.
The first suggested solution was given by Horace Lamb in 1883.2 He postulated that the field was the remnant of some original event and that the decay of this field produced currents which slowed down the decay rate. He developed a rigid mathematical basis for his model using the well-known results of Michael Faraday, the principle of self-induction (inducing a current in a path of good conduction when a field is left to decay).
The mathematical solution to this problem gives an exponential decay but Lamb did not have enough data to evaluate the time constant accurately. Today we have one hundred and thirty years of actual physical measurements (real-time data) and the method of least squares shows a good fit for an exponential decay. The only reason for the rejection of this model by the majority of people today is that the time constant gives an extremely “young” age for the origin of this field. For example, using the evaluation of the time constant by Prof. T.G. Barnes gives a half-life of approximately 1400 years.1 Therefore at about 8000 years B.C. the field would have been that of a magnetic star, at about 50,000 years B.C. the field would have been that of a pulsar and at about 1 million years B.C. the Earth would have been vaporized due to the joule heating of the currents in the core. However, the Earth is assumed to be at least 4.5 x 109 years old and so the majority reject this model.
The dynamo theory
Having rejected the idea that the field is a relic of the past, a means must be found of generating and sustaining it. This is normally proposed as a dynamo action in the core. In the laboratory, models have been built to test the feasibility of the dynamo model but it has not been verified. This is because electrical processes do not scale down in the same way as mechanical processes. Therefore, while it may be disproved in the laboratory, we still cannot say anything on a scale the size of the Earth's core. The dynamo uses these principles:
- Chemical differences set up small currents in the core.
- The motions of molten material give up some of their kinetic energy in amplifying the currents.
- The result is a reasonably constant field being sustained by the kinetic energy of moving particles.
The big question is, are there simple, symmetrical motions which cause the core to act as a self-exciting dynamo? The work of T.G. Cowling3 (1934) shows that no simple, symmetrical motion can produce an exterior field. Nonetheless Elsasser worked out a mathematical description and Bullard explored the possibilities that came forth. These suggestions were all complex motions. Another attempt was noted by J.A. Jacobs who states:
In 1958 G. Backus and A. Herzenberg, working independently, each showed that it is possible to postulate a pattern of motions in a sphere filled with a conducting fluid in such a way that the arrangement acts as a dynamo producing a magnetic field outside of the conductor. In each case the motions were physically very improbable; however, rigorous mathematical solutions were obtained, as was not the case with Ballard's numerical solution. The motions obtained by Backus all involved periods when the fluid is at rest. He needs these periods of rest to ensure that other fields generated by induction will not develop in such a fashion that they eventually destroy the whole process.4
What might cause these fluid motions? There are four possible causes outside of the Earth's gravitational field.
- The bodily tide of the Earth which gives a negligible effect.
- The deceleration of the rate of rotation due to ocean friction which effect is also negligible.
- Precession and Nutation, which cannot be ruled out but are very unsatisfactory and inconclusive at this stage.
- Sudden changes in the rate of rotation. Concerning this last point I quote J.A. Jacobs (emphases mine):
The magnitude of the electromagnetic torque is difficult to estimate but is possibly sufficient—provided the most favorable values of the uncertain parameters are chosen.5
However, the Earth’s gravity may cause motions on inhomogeneities. The first possibility here is chemical differentiation or, that the core is continually growing at the expense of the mantle by the shifting of chemicals. The other possibility is that of thermal convection. Convection currents might be set up by a process of excess heat supply to the core-mantle boundary, excess carried away, or a mixture of the two. Excess heat supplied seems to be ruled out because radioactivity in the core is highly improbable. Concerning the above discussion, Jacobs says that it is not very satisfactory because some values were guesses and many factors were neglected. Jacobs also states:
Items that rather extreme assumptions are necessary to make any theory satisfactory—either an extreme geometry or extreme and implausible values of some of the physical properties in the core and lower mantle. It also appears that the convective heat flow demanded by the theory is excessive and it is not at all certain that the required temperature differences can be realised.6
As of now there is no physical evidence, seismic or otherwise, that there is any motion within the core.7
However, westward drift and secular variation are cited as phenomena which require motions within the core. These effects are basically annual or regular changes in the direction of the field at specific places with time. Concerning westward drift, there is a large area around Canada which is abnormal when compared to the general effect. Barnes suggests that other variations (this might include westward drift) are due to charges emitted by the sun (or solar winds).
To conclude we note that the theory involves much choosing of “extreme” values or ignoring of adverse effects, or very complex and unreasonable motions. Parker says:
So the generation of the field is complicated, but based largely on the streaming and twisting of loops. . . . Calculations show how variation in the strength and direction of cyclonic turbulence in the core of the Earth can lead to active destruction and reversal of the magnetic field.8
This introduces us to the magnetic record believed to be written in the rocks. Many rocks have what is called Natural Remnant Magnetization and this is commonly believed to represent the direction of the Earth’s field at the time of solidification. Therefore, if the dating basis of the geologic column is correct, then the rocks should provide some record of the past history of the field. The strength of the field in the past cannot be measured and this is demonstrated by Chapman and Bartels commenting on E. Thellier’s work:
After reviewing the evidence afforded by his own and other measurements, he concludes that the permanent magnetization of rocks is ill-defined, and gives no safe basis on which conclusions as to the past state of the Earth’s magnetism can be arrived at.9
However, using the directions detected, over thirty reversals of the Earth’s field have been postulated for all of geologic time. Problems arise in the form of some conflicting evidence. Professor L. Neel came up with four possible mechanisms for materials to have their magnetization in the opposite direction to the field, two of which have definitely been verified. This raises the question: what does a reversal mean? Rock reversal or field reversal? J.A. Jacobs says:
To prove that a reversed rock sample has been magnetized by a reversal of the Earth’s field, it is necessary to show that it cannot have been reversed by any physico-chemical process. This is a virtually impossible task. . . .10
However, he also says:
The stratigraphic distribution of normally and reversely magnetized rocks strongly supports the field reversal hypothesis.
To reach this conclusion he dismisses complete self-reversal by viewing that hypothesis through a vast period of time. If the column was laid down at one time, however, much of his reasoning would be invalid.
I would like to return to Horace Lamb’s original model. To do this, all that is necessary is to remove the assumption that the Earth is 4.5 billion years old. There are many things in nature which indicate a very young age for the Earth. They include: population growth, natural gas and oil pressures, lack of meteorites in the geologic column salt concentrations in the oceans, sediment depths on the ocean floor and in deltas, buildup of helium in the atmosphere, breakup of short-term comets, lack of dust on the moon, presence of fine dust in the vicinity of the Earth, rapid breakup of clusters of stars and clusters of galaxies, trilobites and man—contemporary in the fossil record, dinosaurs and man—contemporary, human artifacts embedded in coal, polystrate tree trunks, etc.11 On the basis of this evidence, the good fit of the model with the real-time data, and the extreme complexity of the alternative, I would suggest that the decay model is a more reasonable description of the Earth's magnetic field.
However, more research should be done into western drift and secular variation by those who hold to the decay model. I would also like to see some work done into how reversals fit into the concept of sediments laid at one time by a “flood.”
Let me enlarge on some other problems with the dynamo theory.
- The only complete data we have is real-time data. It is assumed basically, that having started very small, the currents grew until equilibrium was reached. Very little is said concerning the large rate of decrease shown by the real-time data. Elsasser mentions that the rate is slowing down in the hope that the field will begin to increase again,12 but he has assumed that any decrease will be linear and has unwittingly described, at least in form, an exponential decay. The decrease is not trivial, either. Sydney Chapman says that it is a change “not paralleled by any other worldwide geophysical property.”
- Concerning reversals, one method is by rotation, the other is by a magnitude decrease, through zero to opposite sign. The postulated duration is of the order of 10,000 years. Wilbert Rusch asks questions about this event which focus on side effects not normally considered. That is:
- The radiation belts are linked with the magnetic field. What type of storms would have been associated with such a reversal?
- If the field goes through zero and it takes some time to occur, does the midpoint represent a time of no shielding from cosmic rays? Would this not result in the destruction of life on Earth?
- Very importantly—what type of force could reverse the magnetic field?
The only objection to Horace Lamb’s model was the large age assumed for the Earth. Otherwise, not only does his model fit well, but it makes good sense. It is not necessary to assume an old age for the Earth. It is only the prejudice of clinging to uniformitarianism and its offshoots that forces “science” into difficulties in matters like this one.
At the time of writing, Brian D. Eglinton was a 3rd year Engineering student from North Adelaide.