Plate Tectonics

Alan Feuerbacher

The science of plate tectonics is a relatively recent arrival in geology. It came to have wide acceptance among geologists during the 1960s and 1970s, and much information has been added during the 1980s. The August, 1985 National Geographic magazine article "Our Restless Planet Earth" gives a good introduction to the subject. Exploring Our Living Planet63 gives a much more extensive look and is well worth reading. Here I will present much more detail.

The book Islands64 gives a good description of how plate tectonics works:

[The] crust of the earth is a spherical shell of rock that consists of a few rigid plates.... These tectonic plates drift about continually, shifting position and jostling each other. Consequently, the boundaries between them are marked by earthquakes.... Plates are created by solidification of passively upwelling magma, which fills in the cracks where plates drift apart. The cracks are called spreading centers, and they are characterized by tensional earthquakes (caused by stretching), which are confined to shallow depths because the hot crust in these places is too weak for stresses to accumulate any deeper.... Such a crack gradually opened between what are now Africa and South America roughly 200 million years ago. Those continents are far apart now, but the seismically active crack still exists at the crest of the Mid-Atlantic Ridge....

The magnetic field of the earth reverses polarity at intervals on the order of a hundred thousand to a million years. When lava cools, some of the minerals in it act as tiny magnets and orient themselves in the direction of the magnetic field. The spreading crack on the crest of a midocean ridge is frequently filled with lava, which then cools, splits, fills, splits, and so on. Thus the cold rocks of the ridge contain a fairly permanent record, like a magnetic tape recording, of the reversals of the earth's magnetic field through geological time. Indeed, the whole ridge is like a stereo tape recording with magnetic patterns on each flank that are commonly mirror images of each other. The pattern of normal (like now) and reversed magnetic orientations (anomalies) has been dated by comparing rocks of known age on land with those on the sea floor. Inasmuch as most of the magnetic anomalies of the ocean basins have been mapped by ships, the age of most of the vast, deep sea floor is known. Using the width of dated magnetic anomalies, it is possible to measure how rapidly the mid-ocean ridge crest where they were created was spreading apart -- even though it was 100 million years ago.

.... when a spreading center produces an area of new crust, an equal area of old crust must be removed from the earth's surface somewhere.... The regions where tectonic plates drift together and crust is lost are mostly subduction zones. In such a zone, one plate plunges beneath the other, usually at an angle between 30Deg and 45Deg, and goes on down for hundreds of kilometers into the mantle. The plate that plunges is almost always oceanic crust, because continental crust is more buoyant. The path of the plunging plate can be traced by the earthquakes that are generated.... Typically, subduction zones have the largest and most damaging earthquakes in the world because the rocks there are old, cold, and able to accumulate large strains before breaking.

The great compressive forces in subduction zones deform the crust into deep oceanic trenches and high continental mountains such as the Alps and Himalayas. The reheating of the plunging oceanic crust and sediment causes magma to liquefy at depth. It rises to the surface to form lines of beautiful volcanoes like the Cascade Mountains of Oregon and Washington and including the most beautifully symmetrical of all -- Fujiyama in Japan. Like sea-floor spreading, subduction can take place within continents or ocean basins, but in fact it takes place mainly at the boundaries between continents and oceans. The Pacific, unlike the Atlantic, is ringed by subduction zones and the line of fire of active volcanoes. The reason is not that the zones develop at the edges of continents; as at spreading centers, the forces that move plates are much too great to be influenced by the type of crust. What happens is that the buoyant continents drift to subduction zones and stay there like rafts at the edge of a whirlpool in a river.

The crest of the Mid-Atlantic Ridge is not straight; it is offset just like the Atlantic coasts of Africa and south America and for the same reason. The offset is an abrupt step, and it is caused by transform faults, which, like spreading centers and subduction zones, are one of the three basic elements of plate tectonics. A transform fault is, as the name implies, merely a fault, a cut in the earth's crust, running between the other two kinds of tectonic elements. Most transform faults offset the crests of midocean ridges; a few run between a ridge and a subduction zone; even fewer run from one subduction zone to another. The earthquakes on ridge-ridge transforms are quite shallow (1 km to 5 km deep) because the crust there is young and weak. However, where transform faults cut older crust, earthquakes may be 10 km to 20 km deep.... The most famous transform fault is the San Andreas fault....

The earth's rigid surface layer, or lithosphere, is almost wholly in buoyant equilibrium, or isostasy. Because the lithosphere effectively floats on a weak plastic layer, or asthenosphere, its elevation is related to its density. High mountains are composed of rocks of low density, and the deep sea floor is composed of rocks of high density.

The crest of a midocean ridge rises high above the deep sea floor because the young crust created at the spreading center is hot. As the crust drifts away, it cools by conduction to the cold sea floor; it grows denser, so it subsides to form the sloping flanks of the ridge. Cross-sectional profiles of ridges show that their flanks are concave upward between the high crest and the deep basins on either side. It is immediately clear that cooling and subsidence are more rapid when the plate is young than later. The relation between depth and age has been determined in thousands of places and is empirically expressed for crust younger than 60 million years [by a simple mathematical formula].... Thus, it is possible to calculate the expected depth of the sea floor if its age is known. Likewise, the subsidence history of a plate, its depth at any time in the past, can be calculated. The ability to make these calculations has greatly improved understanding of the elevation and subsidence of islands.

The fact that tectonic plates are rigid might seem entirely obvious because solid rock, such as Gibraltar, is the very image of rigidity. However, it is all a matter of scale and the duration of stressing. If the whole earth is shocked almost instantaneously by a great earthquake, it rings like a rigid bell at low frequencies appropriate for its size. On the other hand, geologists have known for a century that the very slow application of pressure to heated rock can cause it to deform like toothpaste or soft clay.... Thus, dealing as they do with millions and billions of years, geologists tend to think of the earth not as rigid but as yielding and plastic.

Moreover, a famous scientific paper in 1937 had demonstrated that it was impossible to lift up a large area of continental crust. (It had a striking cartoon of a giant crane lifting up the earth's crust, 30 km thick, under the state of Texas.) The solid crust proved to be too weak to be lifted up at the edges without sagging in the middle. Consequently, it was a considerable surprise to geologists and geophysicists when it was proved that enormous tectonic plates are rigid. Not in the vertical direction -- plates do bob up and down locally by small amounts to maintain isostasy, and they cannot be lifted any more than Texas can -- but horizontally, they are inflexible.

The rigidity of the plates was demonstrated by appeal to a theorem of the mathematician Leonhard Euler. This states that if one rigid shell moves over another without changing direction, two diametrically opposed points must remain fixed. These points are called Euler poles. The motion of any point on the moving shell may be considered as a rotation around an axis that connects the Euler poles. Relative to the inner shell, points on the moving shell traverse circular arcs centered on the Euler poles. If a tectonic plate is rigid, it can be considered a fragment of a spherical shell (the lithosphere) moving over an inner shell (the earth's mantle). Then its movements must conform to Euler's theorem.

If the Euler poles were, by coincidence, the poles of rotation of the earth, the circular arcs would exactly coincide with the parallels of latitude. In fact, they rarely do, so one must visualize Euler latitudes measured from the actual location of the Euler poles. The motion of a whole rigid plate can be described accurately only by an angular velocity around an Euler pole. The velocity of a given point, however, may be expressed usefully as a linear rate, usually as millimeters per year. This rate varies with Euler latitude from zero at the poles to a maximum at the Euler equator....

A point on the side of a drifting rigid plate, like all other points, must follow a circular arc, and thus the transform fault boundaries at the sides of plates must lie on circular arcs. The crust of the North Pacific was demonstrated to be a rigid plate -- by analysis of the motion on the faults that bound it -- by Dan McKenzie and Robert Parker in 1967. From California to Alaska to Japan, all the faults plotted along circles centered on an Euler pole near Greenland. Jason Morgan, also in 1967, showed that transform faults and fracture zones between plates lie along circular arcs around an Euler pole. Morgan also showed that, in the Atlantic, the widths of magnetic anomalies and thus the rates of spreading vary with the Euler latitude in exact correspondence with Euler's theorem.65

An intriguing discovery in the Pacific was linear chains of volcanic islands, which seemed to continue underwater far beyond the islands themselves. The most striking example is the Hawaiian volcanic chain, which extends from the big island of Hawaii 1000 km west-northwest to Gardiner Pinnacle, through shoals, shallow banks, guyots (drowned ancient volcanic islands that were eroded down to sea level before they sank), and seamounts (drowned volcanic islands that were never eroded), past the island of Midway. It then bends into a north-northwest trending underwater chain called the Emperor Seamounts, which extends to the Kuril Trench off the coast of the Kamchatka Peninsula. The islands farther down the chain give the appearance, and have been measured, to be older than the ones closer to Hawaii. The island chains and the age sequence were first recognized by an American geologist, James Dwight Dana, in the 1840's. Concerning this chain, Islands goes on to say:

Nothing could be more obvious to a geologist now than an age sequence in the Hawaiian Islands. Flying in from the south, one first sees the smooth carapace of the gigantic active volcanoes on the island of Hawaii. Then past Maui to Oahu and Kauai, the islands are ever more deeply eroded into knife-edge ridges and enormous valleys. Assuming only an equal intensity of erosion, the age sequence is manifest. Dana deduced all this at a time when few scientists even believed that valleys are eroded by the streams that run through them. Dana concluded that depth of erosion 'is therefore a mark of time and affords evidence of the most decisive character.'

As sea-floor exploration progressed in the 1950s, the islands tended to be ignored because so many large undersea volcanoes were discovered. These were of two general types: flat-topped guyots, which were drowned ancient islands, and pointed-topped seamounts, which had never been truncated by waves. The guyots were assumed to be geologically old, at least old enough to grow 4 km to 5 km up from the sea floor, be truncated, and sink to various depths. Moreover, some were dredged and proved to be roughly 100 Ma [million years] old. Many of the seamounts and guyots were in lineations trending northwest -- like Dana's island chains with age progressions. However, some groups, such as the Emperor Seamounts, had a distinctively different trend that was almost due north. This was particularly intriguing because the purely submarine Emperor trend was clearly an extension of the Hawaiian trend with a connection at a 'bend' or 'elbow.' Moreover, the Hawaiian trend itself seemed to continue even beyond Midway atoll. Evidently, another stage of submergence carried even atolls beneath the waves....

In 1957 L. J. Chubb.... showed that almost all the high islands in the Pacific have west-northwesterly trends. Not just the ones noted by Dana, but four other groups as well. But atolls are relatively senile chains of islands, and Chubb showed that they have a different trend that is more northwesterly. He would have included most of the submarine volcanoes in his generalization that the 'direction of earth-movements' had changed gradually during geological time from NNW to NW to WNW.

The nature of these 'earth-movements' was proposed by J. Tuzo Wilson in 1963.... He proposed that the sea floor spreads because it is carried along by convection currents that rise from deep in the mantle in thin sheets, flow horizontally beneath the crust, and then return to the mantle in thin sheets. That left a motionless core in each convection cell. Wilson visualized in this core a fixed source of lava, which would rise to build volcanoes. The horizontal convective flow would carry volcanoes away from the source, and the result would be a linear age progression of volcanoes.

But the reality of the age progressions still had to be confirmed. In 1964, Ian McDougall (and later, Brent Dalrymple and others) began to publish the ages of insular volcanic rocks. Dana had been sure only of the sequence in which the volcanoes became inactive. The isotope geochemists showed that each giant island in the Pacific was produced in only a million years or so, and, although there was minor volcanism later, the Dana progression indeed gave the sequence of island formation.

In 1972, Jason Morgan, one of the inventors of plate tectonics, applied it to the problem of the lineations of islands. He showed that the island lineations of the Pacific plot as circular arcs around an Euler pole and that the spacing of islands in the age progressions depends on the Euler latitude. However, the islands do not indicate relative motion between rigid plates but relative motion between one rigid plate and a framework of lava sources called hot spots, in the mantle. If the mantle itself is considered motionless and hot spots do not move around in it, Morgan had tied the past motion of plates into the present geological framework.

Lines of volcanoes have come to be called hot-spot tracks, and they have been studied intensively all over the world. The tracks on different plates can be compared from a knowledge of the relative motion between plates. It appears that hot spots do indeed lie in a fairly rigid framework embedded in the mantle. So-called 'absolute motion' of plates is relative to this framework. Individual hot spots persist for 10 Ma to 100 Ma or more, and, if they drift, it is by no more than a few millimeters per year. Their characteristics indicate that they consist of long, narrow plumes of magma rising from the hot lower mantle. These fixed plumes penetrate the lithosphere and rapidly build volcanoes, which become extinct when drifting separates them from the source area.

Morgan's analysis included the origin of the Emperor-Hawaii bend. Inasmuch as the hot spots are fixed, the Pacific plate simply changed its direction of motion without interrupting the production of volcanoes. From other hot-spot tracks of the same age as the Emperor trend, an older Euler pole was established and with it the motion of all such tracks for about 80 Ma in the Pacific.66

Another type of evidence that led to the acceptance of plate tectonics is related to the apparent wandering of the magnetic poles:

Because the magnetic and geographic poles generally lie fairly close together, determination of the position of the magnetic poles of the past also provides a guide to the probable position of the contemporaneous geographic poles.

Studies of the palaeomagnetism of samples of continental volcanic rocks of different ages have demonstrated that the magnetic poles have changed position over time, appearing to 'wander' widely over the face of the Earth. It is now realised that in most, but not all cases it was the continents, not the magnetic poles, which moved. The rocks of each continent preserve a magnetic record over time of the position of that continent relative to the poles.

For example, the 'polar wander paths' for Africa and South America, plotted independently, follow similar but widely separated tracks... However, if the continents are brought together [with the east coast of S. America abutting the west coast of Africa] the two separate wander paths merge. The obvious conclusion is that for the period represented by the magnetic records (dated radiometrically and/or palaeontologically as covering the period 450-200 million years ago) Africa and South America were closely joined, moved as one unit and began to separate 200 million years ago.67

The above are samples of the tremendous amount of data indicating that plate tectonics correctly describes how the crust of the earth moves, and how mountain ranges and ocean basins are formed. Although many details have yet to be worked out, this does not invalidate the essentials of the theory, any more than my wife's lack of understanding of the chemistry of colloidal suspensions prevents her from making gravy.

Genesis says that the floodwaters were drained within a little over one year after the Flood started, so the events the Society describes as "under the added weight of the water, there was likely a great shifting in the crust.... new mountains were thrust upward.... shallow sea basins were deepened," had to mostly take place in that one year period.

The following sections present what I consider to be the main lines of evidence for plate tectonics. This evidence invalidates the Society's invocation of rapid crustal movements because ".... the continual motion of the plates over the partially molten asthenosphere.... [is what] explains the development of ocean basins and the formation of mountain ranges,"68 and not that "under the added weight of the water, there was likely a great shifting in the crust."69


63 Robert D. Ballard, op cit.

64 H. W. Menard, Islands, Scientific American Books, Inc., New York, 1986.

65 ibid, pp. 27-39.

66 ibid, pp. 41-48.

67 D. R. Selkirk and F. J. Burrows, editors, Confronting Creationism: Defending Darwin, p. 64, New South Wales University Press, Kensington NSW Australia, 1988.

68 Gregory E. Vink, W. Jason Morgan, Peter R. Vogt, "The Earth's Hot Spots," Scientific American, p. 50, New York, April, 1985.

69 Insight On the Scriptures, op cit, Vol. 1, p. 610.