Earthquakes occur everywhere - so, at least, it seems. Temblors happen on all continents and beneath the deep oceans. They shake the world's highest mountains, the Himalayas, and the Earth's deepest valley, the Dead Sea. Even from under the ice caps of both polar regions, seismometers regularly record rumblings in the Earth's crust. But a more detailed look reveals that the distribution of earthquake foci in the world is by no means random. And neither are they evenly or regularly spaced. Instead, when plotted on a world map, earthquake locations look like narrow bands winding through the continents and oceans (see map). What are these zones and why are most earthquakes foci concentrated there?

Simply put, temblors happen when rock breaks under force. Inside the Earth, the most important of such rock crushing forces is the "tectonic stress." It is exerted on the Earth's crust by the movement of the giant, rigid plates, which float on a subterreanean sea of hot and plastic rock called the asthenosphere. There are about twelve huge and another dozen smaller plates. Where ever such plates crush into or slide past each other during their respective drifts on the Earth's surface, the collision is able to break the rock, thus causing earthquakes. In principle, the effects of such plate collisions are similar to a car wreck where two automobiles hit each other, albeit on a much larger scale.

The bands of earthquake foci in the map reflect these collision zones of the tectonic plates. In fact, they very clearly mark the boundaries of the plates. Look for instance at North America. The underlying plate is much bigger than the continent itself. It stretches from Iceland in the East all the way to the most far flung Aleutian islands in the West and reaches from Alaska to the Caribbean and beyond to the Azores, the island archipelago in middle of the Atlantic Ocean.

But earthquakes happen not only where plates collide. They also occur where two plates move away from each other in the so called "spreading zones." One of these zones is the Mid-Atlantic Ridge where Europe moves away from North America at the rate of about one inch per year. You will find such ridges in every major ocean basin. In fact, there are many more miles of plate boundaries under the oceans than on land. As a consequence, the number of submarine earthquakes is also larger than the number of quakes on land. (hra006)

Figure 1: Mud boils caused by liquefaction (From the Karl V. Steinbrugge Collection, Earthquake Engineering Research Center, UC Berkeley). (Click to view larger image.)
Figure 2: Building tilted due to liquefaction caused by the Niigata, Japan, earthquake in 1964 (From the Karl V. Steinbrugge Collection, Earthquake Engineering Research Center, UC Berkeley). (Click to view larger image.)

Everybody has probably done it while frolicking at the beach: Stand with both feet on the wet sand and move your body quickly up and down several times without lifting your feet. After a short while the ground gives way. You sink a few inches into the sand and you might even lose your balance. Well, you just simulated one of the most dangerous effects seismic waves can have on buildings: Soil liquefaction.

How can soil, which is hard enough for you to walk on, lose its strength and stiffness just because it is shaken a little bit? The secret is the water in the soil. Liquefaction occurs only in soils in which the space between individual sand particles is completely filled with water. Such soils are called "water saturated". The water exerts pressure on the particles, which in turn determines how tightly they are packed together. Before an earthquake, the water pressure is low and static. During the dynamic shaking, however, the water pressure can increase so much that the particles can move freely past each other. Once that happens, the soil loses its strength and becomes a gooey, slippery liquid.

"Sand boils" are a relatively harmless consequence of such liquefaction; they look like small mud volcanoes (Figure 1). The overpressure inside the soil causes the sand to squirt out like lava from a volcanic crater. However, when soil liquefaction occurs under a building, it may sink into the soil, like your feet did during your experiment at the beach. The building might even tip over.

The first time seismologists fully recognized the devastating effects of soil liquefaction was in 1964. During an earthquake under Niigata in Japan several apartment buildings sank into the ground and tipped over, because the water saturated soil on which they were built liquefied (Figure 2). The Bay Area is by no means immune to liquefaction, because many buildings in low lying areas are built on soils saturated with water from the Bay. Liquefaction can occur in all areas shaded brown and yellow in the map in Figure 3. The Association of Bay Area Governments (ABAG) has published a booklet with detailed information about the hazards of liquefaction in our region.(hra005)

Figure 3: Map of liquefaction susceptibility in the San Francisco Bay Area (courtesy of USGS). (Click to view larger image.)

	Figure 1: Wreckage of a twenty-one-story, steel-constructed building in the Pina Suarez Apartment Complex. Photo Mehmet Celebi, USGS
	Figure 2: Fifteen-story reinforced concrete structure. Part of the building was only slightly damaged, while another part of it collapsed. Photo Mehmet Celebi, USGS

One of the worst natural disasters in the Americas occured 23 years ago today, when at 7:19 am local time an earthquake of magnitude 8.1 struck in the subduction zone off the west coast of Mexico. The epicenter was located approximately six miles offshore near the town of Zihuatanejo in the state of Michoacan. Although there was severe damage in the coastal regions, the real disaster happened 220 miles away in Mexico City. Less than 15 minutes after the quake, thousands of people in the capital lay dead and the Mexican economy was shattered for years to come. Until today, nobody really knows how many people perished as a result of the earthquake. Official figures for the number of fatalities vary between 9,500 and 35,000. Most people died in Mexico City, where 412 multistory buildings collapsed completely and another 3,124 were seriously damaged, including 13 hospitals. Most of the destroyed structures were between 8 and 18 stories high.

How can an earthquake cause so much damage over 200 miles from its focus? What happened 23 years ago in Mexico is comparable to a temblor occuring along the San Andreas Fault near San Francisco leaving Bakersfield in ruins. To answer this question, we have to go back in history almost 700 years. In 1325 the Aztecs, one of the high civilisations of Mesoamerica, founded their capital Tenochtitlan. They build it on an artificial island in a shallow lake in Mexico's central altiplano. Although the old capital was flooded again and again, the Spanish did not abandon the site in what they called Lago de Texcoco, but enlarged it instead. After Mexico's independence the settlement became the capital of the newly founded country. During the last century, the lake was completely drained, to make room for the housing needs of the ever-growing population of Mexico City.

A lake bed in a basin, however, is one of the worst grounds for constructing a building. While hard rock simply shakes with the same frequency and amplitude as seismic waves, the unconsolidated sediments of an ancient lake bed react differently: They can amplify the shaking and even worse, they can lose their consistency and become a liquid. Such site amplification and liquefaction occured, when the waves of the distant earthquake shook the bed of former Lake Texcoco under Mexico City. Poorly founded multistory buildings lost their footing and collapsed. Read more about the dangers of liquefaction in the next blog entry. (hra004)

In the refrain of a famous German lullaby children are asked: "Do you know how many stars are twinkling in the sky?" The answer, of course, depends on how you look for them and on the brightness of the heavenly objects. Venus, Jupiter and Sirius can be spotted even under bright city lights. If you go out into the country, on a moonless night you can see hundreds of stars with the naked eye and thousands through binoculars. Using the Hubble Space Telecope astronomers are able to spot millions.

The situation is very similar, when you ask how many earthquakes occur during a year. The answer depends on how strong the temblors are, how far away you are from their focus, and how you try to detect them. The shaking of most moderate and all strong earthquakes is so obvious, they they are felt by everybody, sometimes even hundreds of miles from their focus. You may not notice smaller rattlings, say an earthquake of magnitude 4, when you are busily running around. Sitting down quietly at home in Orinda, the blogger has felt even microearthquakes of magnitude 2 occuring almost five miles away on the Hayward Fault under Kensington. However, each year in the Bay Area alone seismologists detect hundreds of earthquakes which are never felt. They use seismometers which are so sensitive, that they pick up the small rumblings of a car driving by hundreds of yards away.

Using networks of such seismometers, scientists have gained a pretty complete picture of how many large earthquake occur worldwide per year (Figure 1). On the average, for every really big shaker of magnitude 8 or larger, there are 17 quakes with magnitudes between 7.0 and 7.9 and 134 temblors with magnitudes in the "sixes". Simply said: With every step down on the magnitude scale, the number of earthquakes worldwide increase by a factor of ten.

Looking at California the earthquake statistic gets somewhat murkier. During the last century we had not a single temblor of magnitude 8 or greater. In the same interval 16 earthquakes occurred with magnitudes in the "sevens" and 39 quakes had magnitudes between 6.0 ad 6.9. For magnitude 5's, the number is in the low hundreds and it reaches just about one thousand for temblors with magnitudes between 4.0 and 4.9. The uncertainty about the number of earthquakes which occur rises significantly as we look for smaller quakes.

As of this writing, more than 400 earthquakes had occured in California during the last week alone (Clickable list of current earthquakes). But only eight of them had magnitudes over 3. The rest were all microearthquakes, which would have passed mostly unnoticed, if it weren't for the more than 600 seismometers, which the California Integrated Seismic Network (CISN) operates in our State. (hra003)

Figure1: Worldwide statistics for large earthquakes in the 1990s (Courtesy of USGS)

Figure 1

Figure 2

The magnitude 4.0 earthquake, which struck under Alamo last Friday night, was felt widely in the greater Bay Area, from Half Moon Bay in the west all the way to Stockton in the east. Many people noticed two distinct jolts, just a few seconds apart. In Orinda, where we were still sitting at the dinner table with family and guests, the two shocks were almost equally strong with about three seconds between them.

Such two separate rumbles are not a sign of two individual earthquakes. Instead they are a consequence of two kinds of seismic waves, which are radiated from every earthquake focus. One type is called a compressional wave, because the seismic energy compresses and releases the ground periodically as the wave passes through. During the shaking, the ground moves in the direction of the propagating wave. (see Fig. 1). The other type is called a shear wave, because it make the ground move perpendicular to the wave direction (see Fig. 2)

on of how these two different waves look.

Compressional waves are also called P-Waves, (P stands for "primary") because they are always the first to arrive. They gave us the first jolt last Friday. Shear waves propagate more slowly through the Earth than compressional waves and arrive second, hence their name S- or secondary waves. They were responsible for the second rumble. The difference in arrival time between these two types of seismic waves can be used as a rough estimate of the distance to the earthquake focus. As a rule of thumb: Multiply the time between the two jolts by 5 and you get the distance to the focus in miles.

Looking at a seismogram of the Alamo earthquake (Fig. 3) the time diffrence between the arrivals of P- and S- waves is 3.5 seconds. This seismogram was recorded at our station BKS, which is located in Strawberry Canyon in the Berkeley Hills above campus. Applying the rule of thumb, we get a distance to the earthquake focus of approximately 17.5 miles. The result is pretty good, as the actual distance is exactly 16.5 miles. (hra002)

Figure 3