UNIT 4 THE LIVING MACHINE : CONTINENTAL TECTONICS AND THE EARTH'S INTERIOR

A. INTRODUCTION

Overview

In this unit you will examine a number of topics in continental tectonics that have been of recent interest. Some are explainable within the framework of plate tectonic theory and others do not appear to be at present. Then, in the second part of the unit, you will look into the deep interior of the Earth, where the driving mechanisms for the plates must surely lie.

B. CONTINENTAL TECTONICS

1. Seismicity Not Associated with Plate Boundaries

Plate tectonics has been remarkably successful in explaining tectonic activity such as earthquakes and volcanoes associated with plate boundaries, but not all geological activity is confined to boundary zones. Earthquakes may be found far from the nearest plate boundaries, as may be seen in Figure 3-2.

Diffuse patterns of earthquakes are found on several continents, notably in China, eastern Africa, and the United States. Seismicity in Tibet and southwestern China is probably related to the ongoing collision between India and Asia. But what of the rest?

HM 4-1: Seismicity in central Asia. Image from NGDC.

In recent years the attention of geologists and geophysicists has turned back toward the continents, seeking an answer to the question of whether this intraplate activity is related to plate motion or whether new and independent mechanisms are at work.

HM 4-2: Red/Blue stereo image: Eastern California:

The long valley running across the lower part of this scene is the Owens Valley of eastern California, a prominent linear fault valley separating the Sierra Nevada Mountains (bottom) from the vast Basin and Range province of Nevada and Utah (top). This rift is part of a general pulling-apart of the western United States occurring in conjunction with the strike-slip movement of western California. Part of Mono Lake is visible at the left edge. The Long Valley volcanic caldera is also visible as a semi-enclosed depression at lower left (centered on the comma-shaped lake). Image and caption from LPI/NASA.

The most widely-felt earthquakes in the United States occurred in the winter of 1811-12 at New Madrid, Missouri, near the junction of the Ohio and Mississippi Rivers. Chimneys fell as far away as Nashville and Cincinnati; church bells rang and pendulum clocks stopped in Charleston and Washington, D.C. The area is still prone to numerous small shocks. In 1886 a major shock hit Charleston, South Carolina, and again was felt throughout a wide region. It would seem that major earthquakes occur less often in the eastern United States, but when they do, they have the potential for much more widespread damage.

HM 4-3 (A-E) Damage from the 1886 Charleston, South Carolina earthquake.

A) The worst earthquake damage in Charleston. Charleston earthquake of August 31, 1886. Charleston County, South Carolina, 1886. (Images and captions from USGS).

B) Collapsed wall units, note the general lack of damage suffered by the wooden structures in the background. These wooden structures were flexible and probably had structural elements which strongly connected different floors. Modern buildings are constructed to flex and bend during earthquake shaking.

C) Close up of failure of wall elements, note the lack of damage to the wooden building in the left background.

D) Train shaken from rails. Note that the rails themselves are unbent, although the ties seemed to have shifted. Compare this with the images of bent rails near Anchorage Alaska damaged by the 1964 Good Friday earthquake.

E) Citizens encamped in city park public square in Charleston after the earthquake. Charleston earthquake of August 31, 1886. Charleston County, South Carolina. 1886.

Why is it that current plate tectonic theory does not provide a ready explanation for these earthquakes? The fact that they occur at all indicates that at least some geological structures within the plates are not fixed and static, but are changing and active. In many cases it would appear that earthquakes occur in regions of the plates that have pre-existing weaknesses that are exploited by large-scale forces acting on the plates. But what is the source of the forces? In its present form, plate tectonics provides an extremely good description for the kinematics of the plates -- that is, how are the plates moving, at what speed and in which direction? The theory has been much less successful in answering questions about the dynamics of the plates -- what are the forces acting within the plates, and what is making them move? Problems of continental tectonics often seem to be related to this latter class of questions, and the possibility of a significant advance in our understanding has many scientists excited.

2. Continental Hot Spots

Yellowstone National Park in the western United States is the scene of active volcanism both now and in the very recent geological past. Geysers, hot springs, and recent lava flows are found in abundance, creating a geological spectacle that draws millions of visitors each year. These features are caused by a large body of hot, low density rock only a few kilometers beneath the Yellowstone caldera, a 70 km (44 mi) wide bowl-shaped depression that includes Old Faithful geyser and much of Yellowstone Lake in the south-central part of the park.

The source of heat may be a magma chamber -- a large room or chamber that has been melted out of the solid rock and completely filled with liquid magma. Crustal fractures give ground water access to this extremely hot region, and the resulting superheated steam powers the geysers, hot springs, and bubbling paint pots that fascinate tourists and scientists alike.

Figure 4-1 maps the location of Yellowstone Park in relation to the Snake River Plain, a swath of basaltic lava extending to the southwest across Idaho. The Snake River plain leads directly into the Yellowstone caldera. To the northeast, beyond the intense activity of Yellowstone, we find nothing. The pattern is similar to that of Hawaii, and Yellowstone is generally regarded as a manifestation of a hot spot.

The arrow in Figure 4-1 shows the direction that the North American Plate is moving in relation to all the major hot spots of the world. We can see, then, that the Snake River plain is the trace of former locations of the Yellowstone hot spot as the North American Plate moves over it to the southwest. It took the Yellowstone hot spot some 15 million years to move from the Oregon border to its present location. Or perhaps we should say that it took 15 million years for the North American Plate to move the distance from the Oregon border to Yellowstone Park. Over the next few million years, we can expect to see the center of volcanism move to the northeast out of the present park boundaries.

Unlike the Hawaiian hot spot, the magma supplying the Yellowstone hot spot must melt its way through thick continental crust before it reaches the surface. The result is a mixture of volcanic rocks ranging in composition from rhyolite to basalt. The brilliant colors of the rocks in the Grand Canyon of the Yellowstone River within the park result from the presence of rhyolite which has been chemically altered by the activity of fumaroles and hot springs.

Fumaroles are vents for hot gases escaping from the ground, while hot springs are conduits for groundwater that has been heated by steam or contact with hot rock far below the surface. Hot water or steam in contact with rocks can chemically alter them in a process known as hydrothermal alteration. Reactions of this kind are of importance in volcanic areas everywhere: at hot spots, subduction zone volcanoes, and at the spreading ridges of the ocean floor. We shall see in the Mineral Resources unit, that these reactions are often important to the accumulation of mineral deposits of economic value.

3. Explosive Volcanism

The Yellowstone caldera had its origin 600,000 years ago in a cataclysmic eruption that must have affected nearly the entire western half of the country. For about a half-million years previous to the eruption, a rhyolitic magma chamber below the caldera had given rise to viscous lava flows extruded through ring-like fractures that developed over the periphery of the chamber.

Magma often has a considerable content of dissolved gases, mostly water vapor. Several kilometers beneath the Earth's surface, the gases remain dissolved in the magma: the pressure is too great to permit the existence of gas bubbles. But if something should occur to lower the ambient pressure below some critical value, then the volatiles begin to come out of solution to form bubbles.

Table 4-1 Recent volcanic eruptions (modified from Christopherson, 1995)
Date Location Approximate number

Of deaths

Volume of extruded material (cubic kilometers)
Prehistoric Yellowstone, Wyoming

Huckleberry Ridge (2.2 million years bp)

Mesa Falls (1.2 million years bp)

Lava Creek (0.6 million years bp)

Unknown

TOTAL 3,580

2,500

280

1,000

4,600 B.C. Mount Mazama (Crater Lake, Oregon) Unknown

75

1,900 BC Mount St. Helens, Washington Unknown

TOTAL 4

Single landslide from north face 2.75

79 Mt. Vesuvius, Italy 20,000

3

1815 Tambora, Indonesia 66,000

150

1883 Krakatoa, Indonesia 36,000

18

1902 Mt. Pelee, Martinique 29,000

Unknown

1912 Mt. Katmai, Alaska Unknown

12

1980 Mt. St. Helens, Washington 72

4

1985 Nevado del Ruiz, Columbia 23,000

1

1991 Mount Unzen, Japan 10

2

1991 Mt. Pinatubo, Philippines 800

12

You can observe this effect every time you open a bottle of soda pop. With the cap in place, the pressure inside the bottle is sufficient to keep most of the carbon dioxide dissolved in the liquid. When the cap is removed, the pressure drops, allowing the carbon dioxide gas to come out of solution in a froth of gas bubbles.

In a magma chamber, the weight of the overburden of rock supplies the pressure that normally prevents gas bubbles from forming. Occasionally, something triggers a pressure drop in the chamber, perhaps a fracture or failure of the chamber roof. The magma froths and expands and begins venting to the surface through any opening that it can find or make. In the same way that a hastily opened bottle of champagne may empty itself of much of its contents, the magma chamber erupts massively, expelling huge quantities of chilled and shattered rock fragments into the atmosphere. With the magma chamber partially emptied, the roof may collapse, forming a giant caldera.

 

The eruption that formed the Yellowstone caldera vented more than 1,000 cubic kilometers (240 cubic miles) of dust and ash that were deposited in a centimeters-thick layer that extended from the Pacific Ocean nearly to the Mississippi River. The ecological effects of such an eruption must be devastating. Total destruction would extend over an area of up to 30,000 square kilometers (12,000 square miles) about the vent, while additional millions of square kilometers would be covered by a blanket of heavy ash. Worldwide climate would probably be affected for sometime by the dust injected into the atmosphere. The total effect on national and global agriculture should such an event occur today would be difficult to assess, but it would likely be severe.

Fortunately, explosive volcanism of this magnitude is rare. Contrast the Yellowstone eruption with that of Mt. St. Helens in 1980, in which less than one cubic kilometer of ash was expelled. Nonetheless, Mt. St. Helens (see location in Figure 4-1) killed 72 people and spread choking clouds of ash eastwards for hundreds of kilometers. Though on a much smaller scale, the explosive eruption of Mt. St. Helens derived its power from the same mechanism of sudden pressure release as the giant caldera events. An earthquake triggered a massive landslide on the north flank of the mountain (Figure 4-2), effectively uncovering the magma chamber. The dissolved volatiles flashed out of solution, unleashing a superheated (300º C) lateral blast moving of speeds up to 250 km/hr, called a pyroclastic surge, that destroyed everything to the north of the mountain to a distance of 21 km (12 mi). A pyroclastic flow is so hot that rock fragments and particles can weld together once again after they have settled to the ground.

HM 4-4 (A-Z5) The eruption of Mt. Saint Helens, May 18, 1980. Skamania County Washington. 

(All images and captions from USGS).

A, B and C: Eruption during the morning of May 18, 1980.

D) Below: July 22, 1980 eruption. Second eruptive pulse approximately 18:45 hours as seen from the south. Note the bicolored appearance of the plume: Lighter color is the ash cloud rising from the pyroclastic flows. Skamania County, Washington.
E) July 22, 1980 third eruption pulse, aerial view taken at 19:07 PDT looking south, showing spreading mushroom top on convectively rising column and cloud of ask rising from an ash flow that has swept northward out of volcano's crater amphitheater. Northwest slope of volcano visible at lower right. Column height about 15 km. Skamania County, Washington. F) Aerial view of eruption of Mount St. Helens volcano from the south. Skamania County, Washington. May 18, 1980.

G and H: Below, before the eruption, Mt. St. Helens was often compared with Mt. Fuji in Japan because of the elegance and beauty of it's form.

I) Mount St. Helens. Blowdown of trees in Green River valley from the May 19 "blast." Cowlitz County, Washington. The direction of pyroclastic flow is exactly marked by the direction in which these dense forests of Douglas Fir trees were toppled. The force of the blast stripped the trunks of branchs. J) Oblique aerial view of more trees downed. Note how blast followed the contours of the mountainside. Skamania and Cowlitz County, Washington. 1980.

K) The force with which the pyroclastic surge struck objects in its path is recorded by the destruction of this tree trunk. L) Tree blowdown, Smith Creek. Two geologists at lower right for scale. Skamania County, Washington. September 24, 1980. The tremendous force of the pyroclastic surge is recorded by the almost complete stripping of the trunks of all of their branches.

M) Aerial view of timber blowdown destroyed by the May 18 eruption of Mount St. Helens. Dirt road for scale. Skamania County, Washington. N) Desolate vehicles in the blast zone created by the May 18 eruption of Mount St. Helens. An ash covered truck and horse trailer near Ryan Lake, more than 12 miles northeast of Mount St. Helens. The vehicles were parked at the edge of the area in which trees were blown down by the lateral blast. Two men were camped at the lake nearby and were asphyxiated by the hot volcanic ash, which covered this spot to an average depth of about 6 inches. The blast temperatures were hot enough to melt the plastic of the truck grill, trailer window, signal indicators, and a lunch box. The tires and glass truck windows, however, were intact. Skamania County, Washington.

O) The high temperature of the pyroclastic surge is shown by these melted gauges. Near view of melted dashboard of pickup truck located on ridge top about 14 km north of Mount St. Helens. Skamania County, Washington. P) A measure of the tremendous force of the pyroclastic surge.Automobile heavily damaged by May 18, 1980 eruption, near Meta Lake, 13 km northeast of Mount St. Helens.

Q) Before the eruption: Warning signs. Fractured bulge on north side of Mt. St. Helens, oblique aerial view. Skamania County, Washington. May 4, 1980. R) Oblique aerial view to the north showing "two-tone" Mount St. Helens - an appearance produced by prevailing easterly winds during the initial activity of Mt. St. Helen. Skamania County, Washington. March 30, 1980.

S) Debris avalanche in the valley of the North Fork Toutle River. View east from near the distal margin of the avalanche toward the cone of Mt. St. Helens, which is partially obscured by clouds in the left background. T) Aerial view east along North Fork Toutle River at Elk Rock bend. Hummocky avalanche deposit of May 18 filled the valley to 45 meters average depth. Mt. St. Helens in background. Skamania County, Washington. June 30, 1980.

U) Geologist lifting bushel basket size pumice rock, near landed helicopter, pumice flow north of amphitheater. Skamania County, Washington. May 23, 1980. V) Oblique aerial view of secondary steam fumarole near Spirit Lake. Fumarole is 300 feet in diameter. Eruptions occurred hourly and continued for two months after May 18, 1980.

W) Devastation along the Cowlitz River resulting from the May 18, 1980 volcanic mudflows (lahars). Garage half-buried by mudflow. Cowlitz County, Washington. X) Newly incised drainage channel in the blast zone, on Smith Creek. In 1980 this drainage channel was approximately 80 feet wide and 20 feet deep. One year later, in 1981, the drainage channel was approximately 200 feet wide and 36 feet deep. Note the remains of trees in the channel and people for scale near center of view. Skamania County, Washington. September 1980.

Y) Destroyed Tower Road bridge, North Fork Toutle River. Cowlitz County, Washington. July 2, 1980. Z) Eastern Washington resident sweeping ash from the May 18, 1980 eruption of Mount St. Helens from the roof of his house. Washington.
Z1) View of Mount St. Helens reflected in Spirit Lake, two years after the eruption of May 18, 1980. Skamania County, Washington. May 19, 1982.  

 
Z2) Aerial view of Mount St. Helens from the northeast before the 1980 eruption activity. Dashed line marks boundary of area removed by the May 18 eruption. Skamania County, Washington. September 18, 1967. Z3) Aerial view of Mount St. Helens from the northeast following the May 18, 1980 eruption. All remaining snow and glacier ice has been covered by ash and other volcanic materials, but the tongue of Forsyth Glacier is clearly discernible below and to the right Dogs Head. Timberline road and parking areas, deeply buried under ash and debris, ate outlined. Skamania County, Washington. July 24, 1980.

Z4) Aerial view of Mount St. Helens from the northeast during the period of deformation and minor eruptions preceding the cataclysmic eruption of May 18. The area of the mountain bulging to the north is outlined. Note the extensive crevassing of glaciers on this bulge. Much of the snow and ice is darken by ash fall. Skamania County, Washington. May 1, 1980. Z5) Aerial view of Mount St. Helens from the northeast before the 1980 eruption activity. Dashed line marks boundary of area removed by the May 18 eruption. Skamania County, Washington. September 18, 1967.

Such lateral blasts can be devastating to inhibited regions in the path of the surge. On May 8, 1902 a pyroclastic surge on the island of Martinique in the West Indies flowed from Mount Pelee covering the town of St. Pierre. Almost the entire population of 28,000 were killed. The high velocity and a cushion of trapped air under the surge allowed it to actually travel over portions of the nearby bay, destroying and setting fire to ships anchored there.

HM 4-5 (A-E): The destroyed city of St. Pierre (1902). Almost 28,000 people were killed when the city was covered by a pyroclastic surge from nearby Mount Pelee, visible in the background. This volcano is located along the subduction zone marking the eastern boundary of the Caribbean plate.

A) View of devastated St. Pierre from slope of Morne d'Orange, looking northeast. Mount Pelee in the distance. Martinique. May 22, 1902. (All images and captions from USGS).

B) This moonscape of devastation was a heavily vegetated tropical region prior to being stripped by the pyroclastic surge. Slope of Morne d'Orange, viewed from the north. May 22, 1902. C) View of debris and devastation on Morne d'Orange. The debris was in part carried up the slope to north and deposited on the top of the hill. Man at left for scale. Martinique. May 22, 1902.
D) The pyroclastic surge leaves a trail of it's direction of motion, here shown by the orientation of fallen trees. Cotton tree and other devastation on Morne d'Orange, blasted by the eruption of May 8, 1902. Martinique. May 22, 1902. E) Guns on Morne d'Orange dismounted by the blast, and other debris from the Mount Pelee eruption of May 8, 1902. Martinique. May 22, 1902.

Another giant volcanic eruption occurred in Long Valley (Figure 4-3) just to the east of the Sierra Nevada Mountains of California some 700,000 years ago. Approximately 170 cubic kilometers of ash were deposited over nine states extending as far east as Kansas and Nebraska.

There have been recent indications that magma is once again on the move beneath the Long Valley caldera. Uplift of the ground surface and a type of seismicity usually associated with magma migration have raised concerns for the population of the area. The town of Mammoth Lakes plays host to tens of thousands of skiers each winter, and substantial efforts have been undertaken to increase monitoring activity so that evacuation can be carried out if significant volcanic activity seems to be imminent. Fortunately, minor eruptions are much more likely than major caldera-forming events.

Mt. St. Helens is a volcano associated with the subduction of a small oceanic plate just off the coast of Oregon and Washington (see the top of Figure 3-13), and Yellowstone is caused by a continental hot spot. The essential cause of the Long Valley volcanism is an accumulation of magma below the caldera, but why is magma being introduced into this region?

Figure 4-3 maps recent volcanism in the western United States. There has been substantial volcanic activity in eastern California, Utah, Arizona, and New Mexico for which plate tectonic theory does not provide a ready explanation. Many of these regions, such as the Rio Grande Rift running north-south in New Mexico, are areas of apparent crustal extension or rifting where the continental crust is being pulled apart and thinned, like a piece of taffy that is being stretched. The extended and thinned crust probably encourages the formation of magma. Why crustal extension is taking place in these regions is not well understood, but probably is related to the release of stress after the cessation of subduction off the coast of southern and central California about 30 million years ago.

NEXT SECTION OF THIS CHAPTER