Topic 14: EARTHQUAKES AND SEISMIC TOOLS Earthquakes - Many people are killed annually during the collapse of buildings or in ground failures caused by the shaking associated with earthquakes. The worst earthquake in the 20th century occurred in Tang Shan, China on July 28, 1976 at 3:42 AM. Of the million inhabitants asleep in the city, 240,000 lost their lives in buildings that collapsed. The buildings were made of unreinforced brick, and could not withstand the shaking associated with a large earthquake. The walls collapsed and the roofs caved in, and the sleeping inhabitants were crushed. In the United States the 1906 San Francisco earthquake is the most famous, but much of the U.S. is affected by seismic hazard. Other notable seismic areas include the subduction zone from northern California to British Columbia (Seattle etc.), Wasatch seismic zone (Salt Lake City), New Madrid (4 Mag 8's in 1811-12!), South Carolina (Charleston), St. Lawrence River-Boston zone. Earthquakes are very important for the analysis of critical facilities, such as nuclear power plants, high level waste isolation, large dams, etc. Seismic loading must be included in the risk analysis of such facilities. Elastic rebound - The earth is an elastic body which is strained during plate tectonic processes. In particular, fault zones at the boundaries of large lithospheric plates can frictionally lock, despite the relative motion of the plates. Over time, large elastic strains can accumulate in rocks surrounding the fault zone, and the shear stress along the fault grows to a large value. When the shear stress reaches a critical value, the locked fault zone ruptures (it "unlocks!") and slip occurs on the fault. This rapid slippage releases in seconds the elastic strain energy that had been accumulated over a long time. As the fault zone slips, seismic waves are released. ------------- Seismic waves - There are two wave types, body waves and surface waves. Body waves travel from the earthquake focus in all directions, whereas surface waves travel along the earth's surface rather than through it. Surface waves are slower than body waves, but cause the most damage. There are two sub-types of body waves, and surface waves, noted below. Wave velocities are defined by elastic properties, and density, and not by the amount of energy released. Compressional (P) waves - body waves characterized by vibrations parallel to the propagation direction of the wave. Because this wave is the fastest of the body waves, it is called the primary wave (P-wave). It typically travels at 5 to 8 km/sec near the surface of the earth and can pass through the liquid core of the earth. The P wave speed for swiss cheese (a lunar material) is about 2.12 km/sec. Shear (S) waves - body waves characterized by a series of sidewise (or shearing) vibrations, perpendicular to the path of wave propagation. These body waves are about half as fast as P-waves, arrive on the seismogram later, and thus are called secondary waves (S-waves). S-wave speed is given by the square root of shear modulus (G) divided by density. As G = zero for a liquid, S-waves cannot pass through the outer core of the earth. Surface waves - two types, Rayleigh and Love waves. Rayleigh waves cause anticlockwise particle motion in a vertical plane. Love waves cause sidewise particle motion, in the horizontal plane. By a property called dispersion, surface waves of different wave lengths travel at different velocities. The surface waves cause most property damage because they cause larger ground displacements, velocites, and accelerations. They also travel more slowly, and may collect wave forms from the entire fault rupture; thus the duration of strong shaking may last for several minutes. The duration of shaking strongly influences ground damage, as shown very well for the Great Alaska earthquake of 1964; there are several big compilations on this very interesting event in the Deike library. longer to pass. --------- Earthquake locations - An earthquake starts at the earthquake focus which is the fault zone along which the earthquake slips. Often the earthquake is located by epicenter or the position on the surface of the earth right above the earthquake focus. Earthquakes are located using a record of the earth vibration called a seismogram. Because earthquake waves travel in at different velocities they arrive at distance seismic stations in a certain order: P-waves before S-waves before surfaces waves. The separation in time of the first arrival of the P- and S-waves correlates directly with the distance of the earthquake from the seismic station and the distance can be graphed on a travel-time curve. For example, an eight minute separation between P- and S-waves on the travel-time curve indicates that the earthquake is 6000 km from the seismic station. Several stations are necessary to pin-point the location of an earthquake. This is done by drawing a circle on the globe with the radius equal to the distance between the epicenter and the seismic station. Having done the same for three stations, the point of intersection of the three circles is the actual epicenter of the earthquake. Earthquake "size" - Several measures exist. One is to record the earthquake's effect on people and buildings. Thus, earthquake "Intensity" is determined on a scale of I to XII (Roman) according to the "modified Mercalli" scale. This method adjusts for different construction standards in different parts of the world. The procedure is useful to show the effects of local geology on ground shaking and damage. It is also useful because it can be applied to historical events, for which no seismograph data are available. A more precise measure of earthquake size is Magnitude, as indicated by the scale devised by Richter, and generally based on surface-wave amplitude. Because the Richter magnitude scale is logarithmic, the difference between two consecutive whole numbers on the scale means an increase in earthquake vibrations of 10 times. Magnitude is a measure of energy release. Earthquake prediction - Because lithospheric plates slip past each other at a known rate, it is possible to determine the frequency with which large faults slip must slip in order to keep up the general motion of lithospheric motion. Because most large faults are locked by friction, seismic slip will release slip accumulated over a long period of time. Depending on the size of the earthquake, slip can be released as often as once every dozen years or as seldom as once every 300 - 400 years (or more). The frequency of slip, and time of the most recent earthquakes, can be determined from detailed inspection of trenches across faults, and age dating of beds cut by faults, and beds that truncate faults. With such information geophysicists can make predictions about (approximately) when the next earthquake could occur along a certain fault. Other methods include the recognition of seismic gaps (spaces on a map or cross-section showing places that have recently moved, and "gaps" indicating places where the fault must play "catch-up" in order to smooth out the slip along the whole plate. The gaps can be quantified, and "conditional probabilities" assigned. A 30% probability was assigned to the Loma Prieta zone on the San Andreas a year before the 1989 earthquake (that caused $5 billion, most costly by far in U.S. history). Still other methods have included monitoring displacements on a geodetic network, tiltmeters, radon gas, seismic velocity ratios, etc. with the idea to be able to predict the coming earthquake very precisely, to the day. This seemed very likely 20 years ago -- researchers were confident that the technology was almost in their grasp. But they are not so confident today as they thought they would be. The problem is still unsolved. Worldwide distribution of earthquakes - Earthquakes can occur almost anywhere in the crust of the earth. However, most of the very large earthquakes occur in several belts which mark the boundary of the large lithospheric plates (which is, in fact, why the boundaries of the plates are drawn there). On a map of worldwide earthquake distribution, the most concentrated belt is the circum-Pacific belt, mostly involving subduction zones and transforms. Another major concentration of earthquakes is in the Mediterranean-Himalayan belt, also generally a place of plate convergence. Shallow focus earthquakes occur along the mid-ocean ridges of the world. Deep focus earthquakes are found in steeply dipping zones called Benioff zones (i.e., seismic subduction zones).