9.2.1 Seafloor topography
The decades after World War II saw a burst of activity in many sciences, including geophysics. Techniques were developed – in
large part from militarysonartechnology, which uses echoes of sound waves to locate underwater structures – to map the ocean floor’s topography, that is, the distribution of elevation caused by undersea mountains and trenches, among other features. The results were striking, as shown in Figure 9.2. The ocean floor is subdivided by long ridges of mountains, stretching over thou- sands of kilometers. These ridges are cut transversely by a series of fractures, orfaults. Near the edges of some continents or major island chains, long trenches are present, the deepest of which is the Marianas trench extending down 11 km from the surface of the ocean, some 7 km below the average seafloor depth.
Even more striking is the distribution of height over the sur- face of Earth, shown in Figure 9.3. Essentially, Earth has two different kinds of surfaces, characterized by their depth: conti- nents and oceans. The range of depths within each kind of sur- face is generally less than the difference in depth between them, leading to abimodal distribution. As we see in Chapter 15, this is dramatically different from the situation on Mars and Venus, suggesting that Earth’s geology has been shaped by a global set of processes unique to our planet. Whether plate tectonics is in fact unique to Earth is explored critically in Chapter 16.
The mid-ocean ridges and trenches are distinguished not merely by their topography, but also by the fact that earthquakes and volcanoes are concentrated along their lengths (Figure 9.4).
Earthquakes also are concentrated along certain fault systems, such as the famous San Andreas fault of California. Because earthquakes are the result of stresses built up by movements in the outer layers of Earth, the presence of narrow belts of activ- ity suggest that the crust is characterized by organized motions rather than by random distortions or deformations.
9.2.2 Magnetic imprints on rocks
Further evidence for moving continents came from an entirely different field of study, calledpaleomagnetism. The magnetic force is one expression of the general electromagnetic force dis- cussed in Chapter 3, which is manifested by charged particles that are in motion. Certain materials, such as the mineralmag- netite, exhibit the ability to retain a permanent magnetization associated with the alignment of the spins of the electrons in the atoms of which they are comprised. Such a magnetic force will cause tiny scraps of iron (iron filings) to line up in a particular manner, which defines the direction of themagnetic fieldof the grains in the mineral (Figure 9.5).
Where does the magnetic field of magnetic or, more precisely, magnetizable, minerals come from? Anyone who has played with a compass is familiar with Earth’s own magnetic field, which, as with a bar-shaped magnet, has a definite directionality to it. A magnetized iron needle suspended so as to rotate freely (i.e., a compass) will point in the general north–south direction – not quite due north–south, because Earth’s magnetic field is not precisely aligned with the geographic poles. The origin of Earth’s magnetic field remains an outstanding puzzle in plane- tary sciences and is discussed further in Chapter 11.
Important to us here is that, as minerals crystallize from molten rock, they do not immediately become capable of hold- ing a magnetic field. This is because at high temperatures the electrons are too mobile to acquire a permanent fixed direction to their spins. Instead, the mineral grains must pass below a
Figure 9.2Map of the Earth’s topography. Beneath the oceans, the seafloor topography is dominated by vast mid-ocean ridges, transform faults, and subduction zones, as described in the text. Red is highest, and blue is lowest, elevation. See color version in plates section.
100
0 0
0
5
HeightDepth HeightDepth
0
5
10
10
5
0
5
10 20
Frequency percent area
sea level
km highest mountain
average height continental
slope sea level
average depth ocean
basin floor
deepest ocean continent
continental shelf
Cumulative frequency percent surface area of world
20 40 60 80 100
50
Area (106 km2) Cumulative area higher than each level (a)
100 150 100 200
(b)
300 400 500 × 106 km2
Figure 9.3Amount of area at various heights and depths on Earth, relative to sea level. (a) Amount is expressed in terms of total area occupied by surfaces with a given height (or depth) in kilometers. The bimodal nature of Earth’s topography is evident. (b) The same data are expressed in terms of cumulative area higher than a given altitude; this gives a sense of the gross profile of continents and ocean floors. From Wyllie (1971) by permission of John Wiley and Sons, Inc.
170 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10
0 180−170−160−150−140−130−120−110−100−90 −80 −70−60 −50 −40 −30 −20 −10 0
0 70
60
50
40 30 20 10 0
−10
−20
−30
−40
−50
−60
−70
70
60
50
40 30 20 10 0
−10
−20
−30
−40
−50
−60
−70 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180−170−160−150−140−130−120−110−100−90 −80 −70−60 −50 −40 −30 −20 −10 0
Figure 9.4Map of earthquake activity over the surface of Earth. Each dot represents a single epicenter. The geographic pattern of volcanic eruptions is very similar. From Isackset al.(1968).
Figure 9.5Appearance of iron filings when a permanent magnet, in this case bar shaped, is placed underneath the glass upon which they rest. Lab set up courtesy of Larry Hoffman, University of Arizona Physics Department.
temperature known as the Curie point to retain a permanent magnetization. This temperature is roughly 800 K and is well below the crystallization point at which the minerals solidify from the melt. Because of this, the direction of the magnetic field inside a permanently magnetic mineral depends on the ori- entation of the mineral, relative to Earth’s magnetic field, at the time it cools through its Curie point.
Hence, if a mineral cools and magnetizes, it becomes a kind of compass. If the rock formation in which the mineral is embed- ded is rotated by 90 degrees, the “north pole” of the magnetic
field in the mineral will point roughly east or west, not north.
This would seem to provide an excellent means for determin- ing the movement of continents relative to Earth’s magnetic field, provided that one can date the rocks by radioisotopic tech- niques. However, interpretation of this information up through the 1950s often focused on the idea that the rotational poles of Earth had shifted with time – either that Earth slowly tumbles through space, or that the entire outer layer of Earth slips over the interior in one piece. In this interpretation, there is no relative drift of the continents but only synchronous changes in latitude around the world.
Studies of continental rocks indicated that rotations in the apparent field directions could not be accounted for solely, or even primarily, by a coherent shifting of all of the continents.
Instead, continental drift had to be invoked to reconcile the directions of remnant magnetization in rocks of various ages and locations. But how were the continents moving relative to each other? The key observation came from the direction of magnetization of rocks on the seafloor.
Study of continental rocks revealed that the Earth’s magnetic field has reversed direction many times in the past – a compass held fixed during a reversal would swing from south to north, or vice versa. These reversals could not be accounted for by a 180-degree rotation of the rock itself, because rocks of the same age from a variety of locations and orientations show the same reversal. The origin of the reversals lies in the way Earth’s magnetic field is generated, deep in its interior, but is as yet very poorly understood. Regardless of their origins, magnetic reversals provide yet another way of delineating the progression of geologic time in the rock record, provided the ages of the
reversals are determined by independent dating of rocks, for example, by radioisotopes.
In the 1950s, technologies developed to detect submarines by their magnetic signatures began to be employed to map the magnetic orientations of seafloor rocks from surface ships.
Very quickly, a remarkable regularity emerged at the mid-ocean ridges, illustrated in Figure 9.6. Stripes of differing magnetic field intensity are laid out parallel to the ridge itself. The vari- ations in intensity are most straightforwardly interpreted to be caused by the direction of Earth’s magnetic field at the time the rock cooled from a magma. The pattern mimics that of the field reversal history recorded in rocks on the continents, indi- cating that the youngest rocks are closest to the mid-ocean ridge, increasing in age farther out. The simplest and most straightfor- ward interpretation of the pattern was advanced in the 1960s: the ridges are sites where new ocean crust is being created, moving like a conveyor belt away from the ridges. As the new crust cools after extrusion, the field direction is recorded in magnetic minerals as they cool below their Curie point. This interpretation quickly led to another question: if new seafloor is being created at ridges, where is it being destroyed?
9.2.3 Geologic record on land
By the 1960s it was becoming increasingly difficult to refute the notion that continents moved about Earth, and that seafloor spreading was somehow involved. Geologic patterns on conti- nents were now being re-evaluated in light of the apparent mobil- ity of Earth’s surface. Radioisotopic dating of similar types of igneous rocks on the eastern end of South America and west- ern end of Africa showed a remarkable correspondence – a well-delineated boundary separating rock of 2-billion-year and 600-million-year ages in western Africa was present in eastern South America as well, and in just the right place for a good jigsaw puzzle fit.
Rocks in mountain ranges in northern California, on the west side of the very active San Andreas fault, matched up well in type and age with rocks a couple of hundred kilometers to the south, in southern California, on theeastside of the fault line.
It was more than tempting to simply slide the east and west portions of California, along the fault, so that these rock types matched. Measurements spanning many decades of ground slip- page resulting from earthquakes showed a northward movement on the western side of the fault of roughly a centimeter per year.
Using that figure, the now-separated rock formations would have been together some tens of millions of years ago.
Drift timescales of centimeters per year seemed to fit in other parts of the world. The magnetic anomaly pattern on the mid- Atlantic ridge displayed ocean floor ages consistent with spread- ing of a few centimeters per year. Studies of the Hawaiian island chain showed that the age of the islands increases to the northwest, with active volcanism confined to the Big Island (Hawaii) on the southeast end and a still-growing submerged island (seamount) just to the southeast of it. If interpreted in terms of the ocean crust drifting northwestward over a source of volcanism, drift rates of some centimeters per year again are obtained.
With mounting evidence in the 1960s for continental drift- ing and seafloor spreading, the question of how the growing
crust was accommodated was answered through the study of earthquakes.
9.2.4 Earthquakes and subduction
Networks of seismometers, which measure the local shaking of the ground due to near and distant earthquakes, are capa- ble of inferring both the geographic location (epicenter) and the approximate depth at which the sudden shifting of rock occurred which caused the quake. (More on this is given in Chapter 11.) Most earthquakes occur at shallow depths beneath Earth’s surface. Careful comparison of Figures 9.3 and 9.4 shows that quakes are common at mid-ocean ridges, trenches, and at sites of lateral movement such as the San Andreas fault in California. However, earthquakes that originate at depths greater than 100 km are confined to the trenches, where some quakes originate as deep as 600 km below Earth’s surface (Figure 9.7).
Earthquakes generally occur where stresses build up in rocks and that stress is relieved suddenly by a failure or fracturing of the rock. Stress build up is generally the result of some move- ment that forces rocks up against each other; in zones such as the San Andreas, such motion is a sliding one in which rocks lock up against each other and then break free. In the trenches, a map of earthquake locations showed an interesting pattern:
the deeper quakes were actually displaced, usually toward the continental side of the trenches (Figure 9.7). The most natural interpretation is that the trenches are regions where ocean floor is sinking underneath continents; as the movement occurs, lock up of rocks causes stress build up and eventual release through fracturing. Thus the trenches are sites at which ocean crust is destroyed: the other end of the conveyor belt that begins at the mid-ocean ridges.
With understanding of the nature of the great seafloor trenches, the basic picture of floating continents was completed in the 1960s and became the theory ofplate tectonics. Tectonics refers to the study of movement and deformation of the outer- most layer of Earth, its crust; plate refers to the emerging concept that Earth’s crust was divided into discrete plates. Seafloor is cre- ated at plate boundaries called mid-ocean ridges and destroyed at subduction zones, which are the ocean trenches. Buoyant continental crust rides passively on the plates, buckling but (for the most part) not subducting when continents on two converg- ing plates collide. This relatively simple model tied together a wealth of geologic data accumulated over the first half of the twentieth century and has shaped our thinking about the geology of Earth and other planets since then.