EARTHQUAKES:

Origins and Predictions

Aspasia Zerva

Department of Civil and Architectural Engineering

Drexel University

Philadelphia, PA 19104

Summer 2000

1

1. Introduction

“Earthquake passed here at 10 p.m. Tuesday … 3 distinct shocks … the first lasted 3 minutes … a sound like a pattering of stocking feet and then the shock that lasted 5 min-utes … then a rolling and rattling sound began … houses swayed badly …’’ (Excerpts from the Palmetto Post, Port Royal, South Carolina in 1886, reprinted in “Low Country Quake Tales” by Joyce B. Bagwell). At 9:50 p.m. on Tuesday, August 31, 1886, a strong earthquake shook Charleston, South Carolina. In 6 to 8 minutes the earthquake was felt as far as Philadelphia, New York City, Chicago and St. Louis. In the Charleston area, 100 people died as a result of the earth-quake, houses cracked and collapsed, sand beds vented explosively into the air and pro-duced craters with surrounding blankets of ejected sand. The earthquake magnitude is estimated between 7.0 and 7.7. Everyday somewhere around the world an earthquake occurs. Most of them are so small, that we cannot even realize that they happened. Some are large and some are great, and, depending on where they occur, they can cause very significant damage. With the catas-trophic earthquakes of last year in Turkey, Taiwan, Mexico and Greece, people started believing that the frequency of earthquakes has increased, and this thought started creat-ing panic. But this is not the case: statistically, the number of earthquakes per year and their magnitude distribution remain the same. Not many people, however, learn about large earthquakes in uninhabited places. On the other hand, our planet becomes more and more densely populated and news and pictures travel nowadays at the speed of light. If a large earthquake strikes in a densely populated area, the damage that it will create will be significant, and will be viewed by people all around the world. In this way, people know earthquakes only from their effects and not from their causes; it is, therefore, important to understand the driving forces behind earthquakes.

2. Origins Undoubtedly, earthquakes are one of the most feared natural hazards. They are unavoid-able, mostly unpredictable, and they produce a feeling of helplessness that cannot be compared with anything else: it is the earth moving under our feet. It is then only natural that our forefathers in seismically active regions around the world, who did not have the information that we have now, believed that it was giant animals or gods who caused the earth to tremble.

Myths: Many myths were created about the possible origin of earthquakes: In Japan, the legend has it that the earth is shaken by the movement of a giant cat-fish hidden in the ground; many Japanese websites still use the word “catfish” in their url address. In China, people believed that the earth is resting on a giant ox, and in India, one myth suggests that the earth was held in place by four giant ele-phants, which were standing on top of a giant turtle, which in turn was standing on top of a giant cobra; whenever one animal moved, the earth was shaking. The

2

ancient Greeks attributed earthquakes to the anger of the gods: when the sea-god Poseidon became angry, he hit the earth with his trident causing the shaking. This fearsome phenomenon was attributed to God’s anger as recently as the 18th cen-tury: people believed that, when society needed chastening, God was sending earthquakes. A Romanian myth suggests that three columns support the earth: Faith, Hope and Charity; whenever people disregarded these virtues, an earth-quake occurred.

However, some scientific explanation of earthquakes can be traced as back as the 5th cen-tury BC: Archelaus, in ancient Greece, attributed earthquakes to compressed air in under-ground caverns – a plausible explanation at the time, as earthquakes and volcanoes are, in many cases, associated, and volcanic eruptions involve the release of compressed gas.

Facts: The facts about the causes of earthquakes were discovered only this last

century, in the 1960’s, with the theory of plate tectonics. The theory suggests that the earth’s crust consists of “plates” that move relative to one another, and seismic activ-ity is associated mostly with this motion. Most of the earthquake sources are located along the boundaries of these plates.

3. Plate Tectonics

In order to understand plate tectonics, it is important first to describe the earth’s interior.

Description of the Earth’s Interior: The earth, with a radius of approximately 6,380 km, consists of three layers: the crust, the mantle and the core, as shown in Figure 1.

Figure 1: Cross-cut along the earth’s interior; the left sketch – shown to scale – indicates how thin the crust is compared to the earth’s size (figure courtesy of USGS, http://pubs.usgs.gov/publications)

3

The uppermost layer, the crust, is only up to 100 km thick: it is as thin as 5 km underneath the oceans, but much denser (heavier) than that underneath the conti-nents, around 30 km underneath the continents, and reaches its maximum thick-ness underneath the large mountain ranges of the earth. Below this thin crust lies the mantle – up to a depth of 2,900 km. The mantle consists of several layers it-self, with its uppermost solid part, that, together with the crust, form the litho-sphere, and, right below it, the semi-solid asthenosphere. The core of the earth consists of a 2,200 km thick liquid layer (outer core), and the center of the earth (inner core), with a radius of 1,300 km, is solid.

The lithosphere is not continuous; it is broken up in irregularly shaped slabs that float on the slowly moving asthenosphere. These irregularly shaped slabs are called tectonic plates, and the theory that describes their motion plate tectonics. Figure 2 presents the seven major tectonic plates of the world, and many of the smaller ones. In many cases the boundaries of the plates are not well defined, as, e.g., between the African and Eurasian plates, where many smaller plates exist.

Figure 2: The major tectonic plates of the world; the shape of the continents is also shown (figure courtesy of USGS, http://pubs.usgs.gov/publications) The development of the theory of plate tectonics is owed to a great degree to Alfred L. Wegener, Harry H. Hess and J. Tuzo Wilson. Their contributions that mark the evolution of our present understanding of plate tectonics are highlighted in the following:

The Theory of Continental Drift: In 1914, Alfred Wegener, a German meteor-ologist, developed his theory of continental drift that formed the basis of plate tec-tonics. He postulated that the earth continents, as we know them now, formed a supercontinent, termed Pangaea (meaning “the entire earth” in Greek), that started to break apart and drift away about 200 million years ago. Figure 3 presents the evolution of the surface of the earth from Pangaea to present. Wegener’s theory

4

was based on the intriguing similarity of the borders of the South American and African continents that appear as if they were fitted together, and the similarity in ancient plant and animal fossils that were discovered in these two continents that are now separated by the Atlantic Ocean.

Figure 3: Evolution of continents on the earth’s surface from Pangaea (about 225-200 million years ago) to present according to the theory of continental drift (fig-ure courtesy of USGS, http://pubs.usgs.gov/publications)

Wegener’s continental drift theory was never accepted during his lifetime. The basic question of the theory’s critics was: “what is the force that is actually driving the motion of these plates?” It took many years and many additional discoveries and developments until the theory of plate tectonics was formalized.

Seafloor spreading: Ocean floor mapping revealed some intriguing features: con-

trary to common beliefs, the ocean floor is very rough with high “ridges” and deep “trenches”. In particular, the mountain ridges seem to surround the continents: In Fig-ure 2, the ridge starts (north) between the North American and Eurasian plates, and continues between the South American and African plate; this ridge is called the Mid-Atlantic ridge. The ridge continues to the east between the African and Australian plates, with a diversion to the north between the African and Arabian and Indian

5

plates, continues further between the Pacific and Antarctic plates and then north again between the Pacific plate and the Nazca, Cocos and part of the Juan de Fuca plates, with diversions north and South of the Nazca plate.

In 1959, Harry Hess, a professor of geology at Princeton University, came up with

the hypothesis of seafloor spreading. He postulated that, along these ridges, the ocean floor is opening up, and magma (molten rock containing minerals and gases), from underneath the lithosphere, is surfacing up and cooling. This process is repeated, so that the ocean floor is moving away from the ridges as new ocean floor is surfacing at the top of these mountains. This hypothesis was confirmed with the discovery of magnetic stripes, parallel to the ridges, that showed magnetic polarity consistent with the reversals of the earth’s magnetic field over the ages, and measurements of the age of the formations, that revealed that their age was becoming younger as their distance from the ridges was becoming shorter.

The question that arises then is: if the ocean floor is spreading, and the earth’s sur-

face size doesn’t change significantly, what happens to the excessive oceanic litho-sphere? The ocean floor mapping revealed not only high ridges, but also very deep trenches. In Figure 2, the plate boundaries between the Pacific plate and the Philip-pine and North American plates are such trenches. There the ocean floor is subducted underneath the continental floor and disappears. As it enters the warmer and deeper parts of the earth, it melts again, so that the oceanic floor is, essentially, recycled. These subduction zones are termed Hadadi-Benioff zones, in honor of the Japanese and U.S. seismologists who first recognized them.

Thus, the theory of plate tectonics was shaped. Still, the question of what is driving the motion of these plates is unanswered. But, it appears that what is happening in the mantle is very similar to what is happening in a pot of boiling liquid: The hotter liquid from the bottom of the pot, where the source of heat is located, moves up, whereas the colder liq-uid from the surface moves down, thus forming convection cells. As the hotter material from the bottom of the mantle moves up towards the lithosphere, the colder material from the surface moves downwards forming convection cells with shape similar to those form-ing in the pot of hot liquid, but of much greater scale. Where the hotter mantle material approaches the earth’s surface, we have the formation of the mid-ocean ridges, and where the colder material starts to move downwards, we have the trenches. The association of plate tectonics and earthquakes is that the vast majority of earthquakes occur along the boundaries of these tectonic plates.

4. Earthquakes and Tectonic Plate Boundaries

The plate boundaries and the earthquakes associated with them are then as follows:

Divergent boundaries: The boundaries along the ocean ridges are called divergent, as the lithosphere is “opening up”. These divergent boundaries are lo-

6

cated mostly in the ocean floor. An exception is the island of Iceland (see Figure 2), where the Mid-Atlantic ridge surfaces up and cuts the island in two. Divergent boundaries are associated with low seismic activity that occurs at shallow depths, because the lithosphere is weak and stresses cannot build up. These boundaries are also associated with volcanic activity.

Convergent boundaries: The plate boundaries where one plate is sub-

merged underneath another or thrust against another are called convergent boundaries. As already indicated, such boundaries exist along the north and northwest border of the Pacific plate (Figure 2). They also exist between the Nazca and South American plates, where the subduction of the Nazca plate re-sulted in the uplift of the South American plate and the formation of the Andes. Convergent boundaries are also associated with volcanic activity. Earthquakes there can be of shallow, intermediate or large depth.

Convergent boundaries also occur between continents. In this case, how-

ever, because the continental lithosphere is light and resists downward motion, the crust is moved either up or sideways. This is the case of the Himalayas and the Tibetan Plateau. The journey of the Indian plate over millions of years can be ob-served in Figure 3: In Permian, India was located right north of what is now Ant-arctica and South-East of Africa. In Cretaceous, India, rotating slightly, has al-ready reached the Equator, and 40-50 million years ago it collided with the Eura-sian plate pushing up the Himalayans and the Tibetan Plateau. These convergent boundaries are also associated with shallow, intermediate or deep earthquakes.

Transform Boundaries: These are plate boundaries, connecting ridges and

trenches, where one plate moves horizontally relative to another. It was J. Tuzo Wilson, a Canadian geophysicist, who first visualized their existence. The most well-known transform boundary is the San Andreas fault in California. The San Andreas fault is part of the boundary between the Pacific and North American plates. The Pacific plate moves northwest relative to the North American plate in the perpetual motion of the earth’s lithosphere, causing offsets of the two sides of the boundary. These boundaries are associated with shallow earthquakes but not associated with volcanic activity.

Plate Boundary Zones: In many cases, the plate boundaries are not as

clearly defined as shown in Figure 2. One such example is the Mediteranean sea, where many microplates form a wide zone. In this case, the African plate is con-verging north underneath the Eurasian plate (and in millions of years the Mediter-anean sea will seize to exist), the Arabian plate is moving north to north-west, and the smaller Anatolian plate (between the African, Arabian and Eurasian plates in Figure 2) is compressed and moves westwards, causing large earthquakes in Greece and Turkey. The large, catastrophic earthquakes in Turkey last year (1999) occurred along the Northern Anatolian fault, which is the boundary be-tween the Anatolian and Eurasian plates. Many smaller plates exist within the

7

area of the Anatolian plate, with complex motions that cause additional seismic activity.

The description of divergent, convergent and transform boundaries clearly defines one type of motion. The boundaries themselves, however, may include more than one form; for example, transform boundaries cross the Mid-Atlantic ridge, which is basically a di-vergent boundary. Although the majority of earthquakes occur along the plate boundaries, there is additional seismic activity within the plates, but it is not yet fully clear why these earthquakes occur. Examples of these intra-plate earthquakes are: The Charleston earthquake, described in the Introduction of this manuscript, occurred within the North American plate, and is thought to be caused by activity in the Mid-Atlantic ridge. There are many small faults underneath Manhattan, some of which may be active; in 1985, a 4.0 magnitude shook the island in New York City. Earthquakes occur in Pennsylvania as well; the Reading, Penn-sylvania, 1994 earthquakes of 4.0 and 4.5 magnitude caused some minor damage. The earthquakes in these areas, as well as those in New England, are thought to occur along ancient, weak zones of plate boundaries. Another example is the three New Madrid, Mis-souri, earthquakes, of estimated magnitude above 8.0, that occurred in 1811 and 1812. These earthquakes were so strong, that formed two lakes (Lake St. Francis and Reelfoot Lake) on the west and east of the Mississippi River, and were felt over an area of 1,000,000 square miles. It is thought that this activity is caused by a system of faults, that are the remnants of an aborted attempt for the formation of a plate boundary in the area. The reason why these earthquakes are felt over large areas has to do with the properties of the rock formations: because earthquakes are not frequent in these regions, the rock formations are not cracked and fractured and, thus, do not absorb much seismic energy. On the other hand, in California, where earthquakes are frequent and the rock formations are cracked and fractured, the seismic energy is absorbed quickly.

5. Earthquakes and Fault Rupture

Let us now formalize the earthquake terminology, which we have, to some degree used already. First of all, fault is a rough surface where shear fracture occurs; the rocks on one or both sides of this fracture surface have slipped along each other. The strike of a fault is the compass direction of its trace on the earth surface, and its dip is the angle (less than or equal to 90o) between the earth surface and the fault surface measured perpen-dicular to the strike. We have encountered the San Andreas fault in California and the Northern Anatolian fault in Turkey. The two sides of these faults move horizontally relative to one another, i.e., parallel to the strike of the fault. These faults are termed strike-slip faults; transverse boundaries are strike-slip faults. The two sides of a fault may also slip along the dip of the fault (dip-slip faults). When the rocks on either side of the fault are pulled apart (ten-sion), as is the case in divergent boundaries, the faults are termed normal faults; shear rupture at normal faults increases the size of the earth crust. When, on the other hand, the

8

rocks on either side of the fault are squeezed (compression), as is the case in convergent boundaries, the faults are termed reverse or thrust faults; shear rupture at reverse faults reduces the size of the earth crust. It is important to note that strike-slip faults may also be associated with some dip-slip activity and vice-versa. Furthermore, it is important to note again, that the plate boundary zones may include more than one form of fault rupture: We have seen that the San Andreas fault is a transform boundary (strike slip fault). However, the 1994 Northridge, California, earthquake occurred along a blind thrust fault, south of the San Andreas fault; the word “blind” indicates that the fault does not break the earth surface. An earthquake then is caused by the sudden slip along a fault. How and why does this happen? We have seen that the tectonic plates move perpetually on top of the aestheno-sphere. This means that the Pacific side of the San Andreas fault in California (Figure 2) moves northwest relative to the North American side. But the rocks adjacent to the fault over extended lengths do not. Why is that? Rock is, basically, elastic material, which means that it can deform under the action of forces, but, when the forces are removed, it returns to its original position. Now, when the plates move, the rock adjacent to the fault cannot: faults are rough surfaces and large forces are required to overcome friction. So the rock adjacent to the fault is stuck, but, further away from the fault, moves along with the plate motion. The transitional part of rock between the “stuck” part and the “moving” part deforms elastically by bending, accumulates strain and stores energy. Eventually, however, enough stress is built up in the rock, so that the friction forces are overcome, and the rock adjacent to the fault rebounds elastically, meaning that it snaps to a position consistent with the plate motion, causing the fault to slip and the earthquake to occur. This, in simple words, is the elastic rebound theory developed by H.F. Reid, after the 1906 San Francisco 8.3 magnitude earthquake. However, not all fault movement results in large earthquakes. At some locations, like at the Calaveras fault near Hollister, Cali-fornia, the fault creeps slowly with time; small earthquakes do occur along these faults, but not major ones, because not enough stress is built up in the material. The location where rupture is initiated on the fault surface is called the hypocenter or fo-cus of the earthquake and its projection on the ground surface the epicenter. The rupture at the fault is a complex process that is very much dictated by the frictional characteris-tics and the shape of the fault surface. The rupture (slip) originates at the hypocenter and may propagate (spread) in one or more directions (unilateral, bilateral rupture), and be arrested in rougher spots along the fault called asperities.

6. What is Felt on the Earth’s Surface

The energy that is released with the rupture at the fault is transmitted throughout the earth in the form of waves. These waves initiate at the hypocenter and start propagating away from the fault, and continue being initiated as rupture spreads along the fault. The process is similar to that of waves forming circles on the surface of the sea when we throw in pebbles, but much more complicated. It is much more complicated because we have many different types of waves and a complex medium – the entire earth.

9

The waves can be classified into two general types: body waves and surface waves. Body waves propagated within the body of the earth, and surface waves along its surface, as depicted in Figure 4. There are two types of body waves: compressional (or primary or P-) waves and shear (or secondary or S-) waves. Compressional waves travel at great speeds and ordinarily reach first the surface of the earth; they displace the earth material particles along their direction of propagation. Shear waves travel slower, but are stronger than the P-waves and displace the earth material particles at right angles to the direction of their propagation. When a P-wave encounters a discontinuity in material properties within the earth, it is reflected back as a P- and an S-wave, and refracted forward as a P- and an S-wave. Similarly, at discontinuities, the S-wave is reflected back as a P- and an S-wave, and refracted forward as a P- and an S-wave. However, only P-waves can propa-gate in the liquid outer core of the earth.

Figure 4: Propagation of waves from the origin of the earthquake through the earth’s body and along its surface(figure courtesy of USGS, http://pubs.usgs.gov/publications) Surface waves propagate on the ground surface. They are also of various types: Rayleigh waves are generated along the earth surface and displace the earth material particles in an elliptical, horizontal and vertical, pattern along their direction of propagation. Love waves are generated at layered media, when the surface layer is of lower density than that beneath it; these waves displace earth material particles parallel to the earth surface and perpendicular to their direction of propagation. A seismogram (recording of the superposition of all the waves) is presented in Figure 5. We can recognize the direct P-wave at the onset of the ground shaking, the arrival of the stronger S-waves, around 8 sec in the plot, and, after these, the longer duration surface waves. This seismogram was recorded by an array of seismographs at the SMART-1 ar-ray (Strong Motion ARray in Taiwan), located in Lotung, Taiwan. The earthquake – called Event 5 – stroke on January 29, 1981, had a magnitude of 6.3 and occurred at a

10

distance of 30 km from Lotung. The seismographs record ground motions in three direc-tions: two horizontal (North-South and East-West) and the vertical. The North-South component of the motions is shown in the figure.

0 5 10 15 20 25 30 35 40−1.2

−0.9

−0.6

−0.3

0

0.3

0.6

0.9

1.2

time (sec)

acce

lerati

on tim

e hist

ory (m

/sec2 )

Figure 5: Accelerogram recorded at station I12 of the SMART-1 array in the North-South direction during Event 5. There are some special features, associated with the earthquake phenomenon, that require special attention. They are:

Site Effects: Sediment basins, depending on their properties, have significant ef-fects on the seismic motions resulting on their surface. These local site conditions tend to amplify the motions, trap waves within the basin and generate long duration surface waves. The effects are, in many cases, devastating. The damage reported in Mexico City during the 8.1 earthquake of 1985 is an example: The earthquake source was some 350 km away from the city, but much of the city is built over the sediments filling an old lake. As a result, the seismic waves incident to this basin were amplified and their duration increased with catastrophic consequences for the city. It was a bad coincidence that the long duration of the seismic waves had a period of 2 sec between wave crests, which was close to the period of oscillation of many multistory struc-tures in the city. When the period of the seismic motions coincides with the period of oscillation of a structure (a phenomenon called resonance), the oscillation of the structure is greatly amplified, so that even well-built structures can collapse.

Liquefaction: Liquefaction is the phenomenon when soil deposits loose their

strength to such degree that they appear to flow as fluids. When liquefaction of soils occurs underneath a structure, the soil can no longer support the structure, and the lat-ter invariably collapses. A spectacular example of liquefaction occurred during the 1964 Niigata earthquake in Japan, that had a magnitude of 7.5. In an apartment build-ing complex, some structures remained intact and standing, some tilted severely from their vertical position, and some, without fully collapsing, tilted so much, that they “lied” on the ground surface. Liquefaction occurs only in saturated soils and is mostly

11

observed near water masses. Associated with the phenomenon of liquefaction is that of lateral spreading, that is characterized by incremental displacements during the seismic shaking. Some dramatic bridge failures occurred due to lateral spreading dur-ing the Good-Friday, 1964 Alaskan earthquake of magnitude 9.2. Landslides can also be caused by liquefaction as well as failure of earth slopes that were marginally stable before the earthquake. Although the scale of landslides is, in many cases, small, it can also be devastating: During the 1970 Peruvian earthquake, an entire village was bur-ied by a landslide that involved 50,000,000 cubic meters and covered, eventually, an area of 8,000 square kilometers; 25,000 people were killed by this landslide.

Tsunamis: A major threat that can be caused by subduction earthquakes is the creation of tsunamis. The sudden motion of the oceanic floor during a major earth-quake displaces large masses of water that travel at the speed of commercial jetliners across the ocean. At open seas the effect of the tsunami wave is not significant. It be-comes however devastating when it reaches shallow waters – i.e., the ocean floor. There the tsunami wave reaches enormous heights and its effect on the coastline is devastating. The islands of Hawaii, in the middle of the Pacific ocean, are particularly susceptible to tsunami waves. Tsunami waves, as well as earthquakes, are also said to be the cause of the destruction of the Minoan civilization in Crete in 1450 B.C.; the waves were created from the eruption of the volcano on the island of Thira (Santorini) and the massive collapse of a large part of the island in the Aegean sea.

7. Earthquake Measurements The size of an earthquake is truly measured by the energy that is released from the fault. However, before modern instruments were developed, the severity of an earthquake was characterized qualitatively from its effects.

Intensity: Intensity is the qualitative measure of earthquake damage and human reaction observed at a specific location. The most commonly used inten-sity scale in the Western world is the Modified Mercalli Intensity scale, developed by the Italian seismologist Giuseppe Mercalli near the end of the 19th century, and was later (1931) modified by Charles Richter in the U.S. The scale, shown in Ta-ble 1, uses roman numerals, so that it is distinguished from the earthquake magni-tude. The numerals range from I (1) to XII (12), with I referring to earthquake events that are “not felt except by a very few under especially favorable circum-stances”, and XII to events where “damage is total; waves seen on ground surface; lines of sight and level distorted; objects thrown upward into the earth”. Addi-tional intensity scales are the Medveded-Spoonheuer-Karnik (MSK) scale used in Central and Eastern Europe, and the one developed by the Japanese Meteorologi-cal Agency (JMA).

12

Table 1: Intensity Scale

The Modified Mercalli Scale of Intensity of Ground Shaking

I Not felt except by very few under especially favorable circumstances.

II Felt only by a few persons at rest, especially on upper floors of buildings.Delicately suspended objects may swing.

III Felt quite noticeably indoors, especially on upper floors of buildings, but manypeople do not recognize it as an earthquake. Standing motor cars may rockslightly. Vibration like passing of truck. Duration estimated.

IV During the day felt indoors by many, outdoors by few. At night, some awakened.Dishes, windows, doors disturbed; walls made cracking sound. Sensation likeheavy truck striking building. Standing motor cars rocked noticeably.

V Felt by nearly everyone; many awakened. Some dishes, windows, etc., broken;a few instances of cracked plaster; unstable objects overturned. Disturbance oftrees, poles, and other tall objects sometimes noticed. Pendulum clocksmay stop.

VI Felt by all; many frightened and run outdoors. Some heavy furniture moved; afew instances of fallen plaster or damaged chimneys. Damage slight.

VII Everybody runs outdoors. Damage negligible in buildings of good design andconstruction; slight to moderate in well-built ordinary structures; considerablein poorly built or badly designed structures; some chimneys broken. Noticedby persons driving motor cars.

VIII Damage slight in specially designed structures; considerable in ordinarysubstantial buildings with partial collapse; great in poorly built structures. Panelwalls thrown out of frame structures. Fall of chimneys, factory stacks, columns,monuments, walls. Heavy furniture overturned. Sand and mud ejected in smallamounts. Changes in well water. Persons driving motor cars disturbed.

IX Damage considerable in specially designed structures; well designed frame structures thrown out of plumb; great in substantial buildings with partial collapse.Buildings shifted off foundations. Ground cracked conspicuously. Underground pipes broken.

X Some well-built wooden structures destroyed; most masonry and frame structuresdestroyed with foundations; ground badly cracked. Rails bent. Landslides considerable from river banks and steep slopes. Shifted sand and mud. Watersplashed (slopped) over banks.

XI Few if any (masonry) structures remain standing. Bridges destroyed. Broad fissures in ground. Underground pipelines completely out of service. Earthslumped and land slipped in soft ground. Rails bent greatly.

XII Damage total. Waves seen on ground surface. Lines of sight and level distorted.Objects thrown upward into the air.

Seismology, however, was revolutionized with the development of the seismograph, the instrument that records the seismic motion, and produces the seismogram, an example of which is presented in Figure 5. The seismograph consists of three basic components: a seismometer, which responds to ground shaking; a chronograph or timing system; and a recording device. Since their initial use in England in 1840 by the English scientist James Forbes, seismographs have been significantly improved.

13

Magnitude: The use of seismographs made possible the development of another, more appropriate scale, to measure earthquakes. This “scale”, termed magnitude, was developed by the American seismologist, Charles Richter, in 1935. In its initial conception, magnitude (which is now known as local magni-tude, ML) is the logarithm (base 10) of the maximum trace amplitude (in microme-ters) recorded on a Wood-Anderson seismograph located 100 km from the epicen-ter of the earthquake. It was found, however, that this magnitude scale was not appropriate for earthquake measurements, because it did not distinguish between the different types of waves that exist in a seismogram. In 1936, Gutenberg and Richter at the California Institute of Technology, presented a world-wide magi-tude scale, the surface wave magnitude, MS, that is based on the amplitudes of Rayleigh waves with period of 20 sec. Later on, in 1945, Gutenberg developed another world-wide magnitude scale, the body mave magnitude, mb, that is based on the amplitudes of the first few cycles of the P-wave. These different magnitude scales produce similar (but not identical) numbers for earthquakes with magni-tudes in the range between 4 and 6, but start deviating significantly for magni-tudes higher than 6, with mb lower than ML, and ML lower than Ms. It is, therefore, important, to know which magnitude is used in the description of an earthquake. An additional problem of the aforementioned magnitude scales is that they be-come “saturated” with the size of the earthquake: as the earthquakes become big-ger, the recorded ground motion characteristics become less sensitive to the earth-quake size. Hiroo Kanamori at Caltech in 1977, and a few years later, Kanamori and Thomas Hanks developed the moment magnitude scale, Mw, that does not be-come saturated with earthquake size. This magnitude is directly related to the source rupture by means of the following equations:

7.105.1

log−= O

wMM

where M0 is the seismic moment given in (dyne – cm) by the following equation:

DAM µ=0

withµ being the rupture strength of the material along the fault, A the rupture area and D the average amount of slip. The total seismic energy released during an earthquake, E, is estimated from a relation derived by Gutenberg and Richter, and later shown by Kanamori to be valid for the moment magnitude as well; it is:

SME 5.18.11log += where E is expressed in ergs. This relation shows that if Ms is increased by one unit, then the released seismic energy is increased by 101.5, which means that the energy released by an earthquake of Ms =6 is abut 32 times bigger than the one re-

14

leased by an earthquake of Ms =5, and about 1000 times bigger than the one re-leased by an event with Ms =4.

8. Predictions

Given that we have all the aforementioned information regarding plate tectonics, the most important question that arises is whether we can predict where and when the next earth-quake will occur, so that people can be warned and loss of life and property can be re-duced to a minimum. Presently, the available approaches for estimating the potential of the occurrence of an earthquake at a site are based on probabilistic methods and strain measurements. The probabilistic methods are based on historical evidence of large earthquakes in a region; if, for example, five earthquakes of magnitude 6 have occurred in the last 200 years (or every 200 years in known times) at a fault, then it is reasonable to assume that a 6 magni-tude earthquake will occur in the next forty years, i.e., its return period is forty years. A problem with this “prediction” is that fault systems are interrelated: when strain is re-leased at one part of the fault, it may increase or decrease elsewhere, so that the strict probabilistic evaluation of the return period may not be valid. Strain measurements are also performed along faults, and, in particular, the San Andreas fault. The evaluation of strain measurements, how much time has passed since the last earthquake and how much energy was released during the last event give scientists an indication of when the next event is going to take place. These evaluations, however, are still subject to the interac-tions of the particular fault with additional faults in its vicinity, and, also, these elaborate measurements and evaluations are not performed on every known fault. Additional ap-proaches for the prediction of earthquakes have been proposed by scientists around the world, but they are still in their development and justification stages. The safest approach, for the time being, for the reduction of loss of life and property dur-ing an earthquake is the evaluation of the proper seismic hazard for the region, and the design and construction of structures capable to withstand this hazard. Seismic hazard is the quantitative description of ground shaking caused at a particular site, and can be ana-lyzed deterministically or probabilistically. The probabilistic approach considers uncer-tainties in the earthquake size, the fault, where the earthquake will occur, and the return period of earthquakes, among other parameters. The deterministic analysis considers a particular earthquake scenario at the site of interest. In this case, the controlling earth-quake (the one that will produce the strongest shaking at the site) is identified from all existing faults and their shortest distance to the site. Based on the controlling earthquake, structural design parameters are identified, and structures are designed and built. Addi-tional considerations in seismic hazard evaluations are the special features of the earth-quake phenomenon, such as the site effects highlighted earlier.

15

9. Conclusions We have learned what are the driving forces behind earthquakes, how tectonic plates move on the earth’s hotter interior, what their boundaries are, and how the rupture at the fault causes an earthquake. We have also seen that, what we know now, is the accumu-lated effort of scientists over the last thirty to forty years, which is a miniscule amount of time compared to the amount of time that was required to make the earth surface be what it is now. Actual seismic prediction, however, is yet to come; let us hope that its achievement will be feasible and in the not-too-far future.

10. Assignment A. Select an earthquake from the following regions of the earth:

U.S. (Alaska, California (North), California (South), Central US, Hawaii, Penn-sylvania, South Carolina), Chile, China, Greece, Iceland, Italy, Japan, Mexico, New Zealand, Peru, Taiwan, Turkey.

B. Fully document the earthquake:

1. Location, faulting, and magnitude 2. Damage description

C. Associate the earthquake with historic seismic activity in the area and plate tectonics. D. Recommend seismic hazard for your selected area. Additional Reading / Bibliography D.S. Brumbaum, Earthquakes. Science and Society, Prentice Hall, NJ, 1999. S.L. Kramer, Geotechnical Earthquake Engineering, Prentice Hall, NJ, 1996. C.F. Richter, Elementary Seismology, W.H. Freeman, San Francisco, CA, 1958. W.J. Kious and R.I. Tilling, This Dynamic Earth: Theory of Plate Tectonics, U.S. De-partment of the Interior/ U.S. Geologic Survey. U.S. Geologic Survey, http://pubs.usgs.gov/publications.

16