1

Kahua A‘o: Plate Tectonics

Plate Tectonics has revolutionized geological thinking since the early 1960s. In essence Plate Tectonics says 2 things:

1. THE EARTH’S SURFACE IS DIVIDED INTO A NUMBER OF PIECES (THE PLATES), AND...

2. THE PLATES MOVE

--First, a little history---

Since geology started being serious, geologists have been looking for ways to explain a number of things, including:

1. The bi-modal distribution of the Earth’s surface (oceans and continents).

2. The spatial and depth distribution of Earthquakes.

3. The distribution of Volcanoes.

4. Large-scale features such as mountain ranges and rift valleys.

5. The linearity of many mountain ranges.

6. The distribution of similar fossils, rocks, and geological structures in places that are far apart.

In the 19th century, geologists noticed crumpled mountain ranges such as the Appalachians and figured that as the Earth has cooled it has contracted, and its skin is getting wrinkled. BUT…if the Earth is contracting, how could there also be areas of obvious tension such as the East African Rift and the Basin & Range in SW N. America??

Fig. 1, top, showing how a shrinking planet could produce crumpled zones where two thick segments collide during shrinkage. Middle image shows a side view of a graben. Bottom shows how an expanding planet that started out as one single continent, would eventually develop multiple continents.

Then, when radioactivity was discovered, geologists decided that this internal heat source meant that the Earth wasn’t simply radiating away its primordial heat, it was also producing heat. The calculated cooling rate (incorrect, as it turned out) was lower than internal heating rate if radioactivity-produced heat was factored in. So rather than shrinking, maybe the Earth is actually expanding! Perhaps Earth started out as all one continent. As the planet expanded, that one continent would have to rift apart (to produce rift valleys), and perhaps basalt could well up into the cracks (to produce the oceanic crust). This would explain how the pieces can be fit back together like a puzzle. However, this makes explaining the crumpled-up mountain ranges, which were obviously produced by collision, kind of difficult. The problem, obviously, is that there are areas that show clear evidence of previous collisions and other areas that show clear evidence of having been pulled a part. Any explanation that produces only contraction or only expansion isn’t going to solve the problem.

Along came ALFRED VEGENER, a German meteorologist, who, in 1912, suggested that the continents drift slowly across the Earth’s surface, sometimes breaking apart, other times colliding. He likened them to Ships of Stone sailing through an ocean of stone. The main selling point of his theory was that there would be both tension and compression: compression at the leading edge of a plate, tension at the trailing edge. Wegener was mainly trying to explain the fit between South America and Africa. He was not the first to puzzle over this, however: Francis Bacon noted the fit in 1620, and Antonio Snider, in 1858, published maps depicting continental drift and likened the process to breaking up a sheet of ice on a pond. If all this were true, THE PIECES SHOULD FIT BACK TOGETHER AGAIN. We’ll come back to this in a bit, but first, let’s take a break to see what these “plates” are, actually.

Figure 2. The compositional (left) and physical (right) layers of the Earth.

The Earth can be divided into layers, sort of like a great big onion. There are two main ways in which the layers are distinguished: 1) by their chemical properties; and 2) by their physical properties. Both schemes are equally valid, but if the terms are used carelessly (which is often the case), things can get confusing. Neither of the schemes for describing the Earth’s layers is better than the other – which one you use depends on what you are trying to discuss. An analogy might be a classroom. The teacher can divide the students by gender, hair color, age, hometown, etc. All of these classifications will operate on the same set of students, and none are inherently better than the others.

THE CHEMICAL LAYERS

CRUST: comes in 2 flavors, continental and oceanic.

Continental crust isthick (30-60 km), has alower density (about 2000 kg/m3), and is rich in minerals that are silicates of aluminum, calcium, potassium, and sodium, such as feldspars and quartz. Many intro-geology textbooks say that the average composition of the continental crust is granodiorite (a rock sort of like granite), but this often confuses students into thinking that the continental crust is granodiorite. It is probably better to say that continental crust is very diverse and includes pretty much every rock type that exists, from basalt to granite to limestone to sandstone to metamorphic rocks, and so on.

Oceanic crust isthin (2-5 km), has ahigher density (about 2500 kg/m3), and is rich in minerals that are silicates of magnesium and iron, such as olivine and pyroxene. The average composition of oceanic crust is basalt. This is much more realistic because much of the oceanic crust really is basalt or its intrusive equivalent, gabbro.

MANTLE: The mantle has af higher densitythan the crust (about 3300 kg/m3), and is composed mostly of olivine and pyroxene (or their high-pressure equivalents). A rock consisting of olivine and pyroxene is called peridotite. No drill holes have ever reached the mantle, although people have tried. There are a few slivers of the uppermost mantle that have been slapped up onto the surface where plate collisions occurred and thereby exposed on the surface. Such rocks are called ophiolites. Additionally,some volcanoes occasionally bring up pieces of the mantle as inclusions in their erupting magma. Such inclusions, whether they come from the mantle or crust, are called xenoliths (foreign rocks). Most of what we think we know about the mantle comes from indirect study (earthquakes). A very prominent change in the propagation speed of earthquake waves (from slower to faster) called the Mohorovicic Discontinuity (usually called the Moho) defines the crust-mantle boundary.

CORE: The core is the densest layer, and it consists of an iron-nickel alloy, perhaps with a little bit of sulfur mixed in. We cannot sample it, and no magma sources are that deep (so no volcanoes bring up core xenoliths), so all evidence is indirect:

1. Many meteorites consist of iron-nickel alloys. Most meteorites are thought to be fragments of a broken-up planet that was once like Earth. We don’t find lumps of iron-nickel alloy lying around on the sEarth’s urface so if Earth is indeed like that broken-up planet, our Fe and Ni must be down in the core. Additonally, we generally know the elemental abundance of the inner solar system, and because the crust and mantle combined do not contain sufficient Fe to match the solar system abundance, Earth’s Fe (and Ni) must be in the core.

2. From studies of orbital mechanics involving the moon and sun we know that the average density of the Earth is about 5500 kg/m3. Crust and mantle rocks are both less dense than this so there must be something denser to make up the difference:

densities of...

Earth = 5500 kg/m3

crust = 2000-2500 kg/m3

mantle = 3000-3500 kg/m3

FeNi meteorites = 7000 kg/m3 but under core pressures = 10,000-12,000 kg/m3

People sometimes wonder why mantle convection can’t drag up pieces of the core from the core-mantle boundary. What they don’t realize is that the density difference between the core and mantle is 3x greater than the density difference between the crust and atmosphere. How often does convection in the atmosphere drag up pieces of the crust?

Thus the Earth is chemically zoned. 90% of the whole Earth is Fe, O, Si, and Mg, but 90% of the crust is O, Si, Al, and Fe. The elements get layered but because they form compounds they are not layered simply by their atomic weights. For example, uranium, lead, and thorium, which are dense elements, readily form low-density compounds with oxygen and are thus concentrated in the continental crust.

THE PHYSICAL LAYERS

Based on evidence from earthquakes, gravity surveys, and lab experiments, we can also divide the Earth into physical layers:

LITHOSPHERE: “rock layer”: This layer includes the crust and the uppermost mantle. It is cold, rigid, and brittle. It extends down to about 100 km. The “plates” of plate tectonics are pieces of lithosphere. Therefore we should avoid the term “crustal plates”, and instead use “lithospheric plates”.

ASTHENOSPHERE: “weak layer”: This layer is part of the upper mantle. Although solid, is hot, weak, and ductile (able to deform). Deformation of the asthenosphere allows the tectonic plates to move.

MESOSPHERE: “middle layer” This is most of the mantle. It is hotter than the asthenosphere, but because the pressure is higher, it is a much less ductilesolid.

OUTER CORE: Seismic evidence shows that this layer of iron-nickel alloy is liquid. Thermal convection and rotation of the Earth cause the outer core to move. It is therefore a moving electrical conductor, and this movement produces a magnetic field, namely the Earth’s magnetic field.

INNER CORE: This layer is also iron-nickel alloy, and although even hotter than the outer core, the high pressure keeps it solid.

Where is all this heat coming from??? What drives plate tectonics? Where does the heat for producing volcanoes come from? The two main sources of heat are decay of naturally-occurring radioactive elements, and heat leftover from Earth’s formation. Even though Earth formed 4.6 billion years ago, rock is a very good insulator so it isn’t unreasonable for there to still be some heat left. A tiny amount of heat energy comes from meteorites striking the Earth, but it is so tiny that it can be ignored.

OK, back to plate tectonics...

Some pieces fit together very easily, such as the east side of South America fitting into the west coast of Africa, especially when you fit the continental shelves rather than the coastlines. If you put these pieces back together you can explain a bunch of things:

1. IdenticalLAND fossils on both sides of the Atlantic (animals that didn’t appear to have been swimmers based on fossil evidence would otherwise have had to swim from one side to another).

2. Identical TREE fossils in many places (trees are even worse swimmers than big animals). The fossils of one particular tree, Glossopteris flora, can be found in India, S. America, and Australia. Glossopteris fossils older than ~225 million years are pretty much the same in all three places, but from that age foreword they are more and more different. If, prior to ~225 million years ago (a geologist would say ~225 Ma), India, S. America, and Australia were connected, it would make sense that the same trees might be found on all three. However, if around 225 Ma the continents started to drift apart into different latitudes, it also makes sense that the Glossopteris trees would start to evolve differently.

3. Coal forms from what were once tropical rain forests and swamps, but there are coal deposits in what are now cold places (Antarctica, N. America, Europe). If these continents were once closer to the equator where the climate is warmer, warm swamps could have existed.

4. There is evidence that some 300 million years ago a single glacial ice sheet covered parts of S. Africa, S. America, India, and Australia. By looking at scrape marks in the rocks caused by the moving ice (called glacial striations), the directions of the ice movement can be determined. If a single ice sheet covered all these places in their present positions it would have been huge. Additionally, in their present locations, most of Africa, S. America, India, and Australia are too warm to be glaciated, even during ice ages. Finally, there is no evidence that there were huge glaciers in the N. hemisphere at that time. But…if all the southern continents were joined together and in a location much farther south than they are today, all these problems can be solved.

Figure 3. Top: The southern continents in their present positions, showing the extent of ~300 million year-old glacial deposits. Bottom: Reconfiguration of the southern continents, producing a much more reasonable ice sheet, with glacial striations indicating radially-outward motion.

The problem was (and still is, to some extent)...HOW DO THEY MOVE?

Wegener said they slide or “sail” through the oceanic crust, but there are no earthquakes to indicate that continents are plowing through solid rock.

Holmes (1928) said that there might be convection in the asthenosphere and like a conveyor belt, the continents can get dragged/carried along, break up, or collide.

By the 1930s (Wegener died in 1930), the debate had become a stalemate with the “drifters” looking for a way to move the continents and the “anti-drifters” looking for ways to explain many geologic observations without moving the continents.

THEN, in the 1950s, largely due to oceanographic work begun during World War 2, scientists started making great strides into two relatively-uncharted areas: The OCEANS and PALEOMAGNETISM, the study of the Earth’s magnetic field.

Oceanographic studies, which were started during the war in an effort to find places to look for, and/or hide, submarines, came up with the following findings:

1. There is a 65,000 km-long, 2000 km-wide, 3 km-high mountain range snaking through the central parts of most of the oceans, kind of like the stitches on a baseball. This was named the mid-ocean ridge.

2. There is a graben at the summit of this ridge indicating that it is a zone of tension.

3. Ocean sediments get thicker the farther away from the crest of the ridge you are.

4. There is a high amount of heat escaping to the surface along this ridge.

5. There are lots of shallow earthquakes along this ridge.

6. There are lots of deep earthquakes along many of the oceanic margins where the deep trenches are.

7. Rocks dredged from the summits of the mid-ocean ridges are volcanic.

8. The oldest sediments anywhere in the oceans are only ~200 million years old.

Figure 4. Diagrams showing the different results for true polar wander vs. apparent polar wander (caused by plate wander).

9. There are areas of anomalously high gravity over the ridges and of anomalously low gravity over the trenches.

Paleomagnetic studies came up with interesting discoveries as well. Rocks have weak magnetic fields. When a volcanic rock containing the mineral magnetite cools,the magnetic field of all the magnetite grains line up in a direction parallel to the direction of the Earth’s magnetic field. This means that a stack of rocks, accumulated over time, acts like a tape recorder showing the direction from the rocks in each layer to the magnetic pole.

If you go to a place and measure the orientation of the rocks’ magnetic fields and find that they are different from the present magnetic field, either the magnetic poles have moved relative to the rock or the rock has moved relative to the magnetic poles. Or both. How do you decide which happened? Plot the apparent pole positions of two different continents through time. If the paths are the same (i.e., the magnetic field in the 150 million year-old rocks of Continent A points to the same location as the magnetic field in the 150 million year-old rocks of Continent B, then the pole has been moving and the continents have remained fixed. If the paths are different, however, then the pole has remained fixed and the continents have moved.

Figure 5. Diagrams showing how sediments are thicker the farther you go from the central graben of a mid-ocean ridge (top), how a stationary magma source and moving plate can produce a chain of ever-older volcanoes (middle), and how bands of normally- and reversely- magnetized rocks produce symmetric stripes centered on a mid-ocean ridge.

The magnetic field does drift around a little (called secular variation), but most of what is sometimes called “apparent polar wander” is actually due to the continents moving, not the magnetic poles.

In 1962, Harry Hess proposed SEA FLOOR SPREADING, and subtitled his paper “An Essay on Geo-poetry”. This idea said that magma rises from depth beneath a mid-ocean ridge and pushes its way to the surface. In so doing it pushes the existing rocks to the side. The next batch of magma then pushes itself to the surface, and so on. This would mean that the farther you were away from the ridge crest, the older the crust would be and the longer it would have had to sit around and collect sediments. One problem with this idea is that if the plates are pushed apart, there should be indications of compression acting on the two plates. Instead, there is evidence of just the opposite, namely tension, indicated by the graben running along the crest of every mid-ocean ridge.

J. Tuzo Wilson suggested that if you can find a long-lived magma source that didn’t move (maybe because it was below the moving plates), but produced volcanic islands high enough to break sea level then as the crust moves over it a linear chain of volcanoes will be produced. These volcanoes will be young at the “upstream” end and old at the “downstream” end. Such an unmoving magma source is now called a “hotspot” although “hot column” would probably be more appropriate.