Throw a rock into a pond or lake and watch the waves rippling out in all directions from the point of impact. Just as this impact sets waves in motion on a quiet pond, so an earthquake generates seismic waves that radiate out through the Earth.

Seismic waves lose much of their energy in traveling over great distances. But sensitive detectors (seismometers) can record theses waves emitted by even the smallest earthquakes. When these detectors are connected to a system that produces a permanent recording, they are called seismographs.

There are many different types seismometers, but they all are based on the fundamental principle - that the differential motion between a free mass (which tends to remain at rest) and a supporting structure anchored in the ground (which moves with the vibrating Earth) can be used to record seismic waves.

Figure 1. Seismographs are designed so that slight earth vibrations move the instruments; the suspended mass (M), however, tends to remain at rest, and its recording stylus records this difference in motion. The horizontal seismograph shown here moves only in the horizontal plane. Vertical seismographs, like the simple one shown here, use a "soft" link between the earth-anchored instrument and the suspended mass. In this design, the mass hangs from a spring, which absorbs some of the motion and causes the mass to lag behind actual motion.

This principle is illustrated in figure. Vertical support AB holds mass M in position by wire AM and by strut BM at point B; the system becomes a seismometer when the vertical support is embedded in a concrete pier attached to the Earth. If there is no friction at the point B and mass M is reasonably large, the movement of the pier and the attached upright support in response to an earthquake wave will set up a differential motion between the mass and the pier (the inertia of the mass will make it remain at rest). This motion - the signal of an earthquake wave - can then be recorded on a revolving drum. When the pier is steady, the pen attached to the mass writes a straight line. But when the pier shakes, the mass and strut wiggle, recording waves from the earthquake that started the boom in motion.

Usually, the drum rotates on a screw-threaded axle so that the recording pen moves on a continuously advancing record and does not simply repeat the same circle over and over. Because time - both the time of day and the synchronization of events - is an important element in seismology, clocks are always part of a seismograph system.

A single seismograph pendulum works in only one direction, and cannot give a complete picture of wave motions from other directions. To overcome this problem, modern seismograph stations have three separate instruments to record horizontal waves - (1) one to record the north-south waves, (2) another to record east-west waves, and (3) a vertical one in which a weight resting on a spring tends to stand still and record vertical ground motions. The spring-suspended mass lags behind the motion caused by the earthquake, making the pen record the waves on the drum. This combination of instruments tells a seismologist the general direction of the seismic wave source, the magnitude at its source, and the character of the wave motion. Instruments at other stations must be used to get a precise fix on the earthquake's epicenter.

An earthquake generates a series of waves that penetrate the entire Earth and travel at and through its surface. Each wave has a characteristic time: each has its own move of travel. They are quite complex, but a few basic facts will explain how they travel through the Earth and how an earthquake's epicenter can be determined from seismograph records.

There are four basic types of seismic waves; two preliminary body waves that travel through the Earth and two that travel only at the surface (L waves). Combinations, reflections, and diffractions produce an infinity of other types, but body waves are the main interest in this discussion.

Body waves are composed of two principal types; the P (primary) wave, comparable to sound waves, which compresses and dilates the rock as it travels forward through the Earth; and the S (secondary) wave, which shakes the rock sideways as it advances at barely more than half the P-wave speed.

Figure 2. Travel-time curves with idealized seismograms (earthquake records superimposed).

The P wave is designated the primary preliminary wave because it is the first to arrive at a seismic station after an earthquake. It travels at a speed usually less than 6 kilometers per second in the Earth's crust and jumps to 13 kilometers per second through the core.

The S wave is the secondary preliminary wave to be recorded. It follows paths through the Earth quite similar to those of the P-wave paths, except that no consistent evidence has yet been found that the S wave penetrates the Earth's core.

The lines labeled P, S, and L in the curves shown on figure 2 represent the travel time required for each phase at distances of 0 to 1300 kilometers from the earthquake's epicenter. They mark the points on the record at which these waves first arrive at the station.

The simplest method of locating an earthquake on a globe is to find the time interval between the P- and S-wave arrivals at several seismograph stations. The distance to the earthquake from each station is then determined from standard travel-time tables and travel-time curves. Great-circle arcs are drawn on the globe using the distance of the earthquake to the station as a radius. All the arcs should intersect at a common point - the epicenter.

Another method of locating an earthquake is to use the P-wave arrival-time minus origin-time (P - O) interval instead of distance. This method is more common because the time can be taken directly from surface focus travel-time tables assuming an origin of 00 hours. This method, however, requires that travel-time tables be available for various depths of focus. For locating a deep shock, one 700 kilometers deep, for example, travel-time tables and travel-time curves for that depth have to be used to calculate the origin time and distances.

Other wave types can be generated inside the Earth by P and S waves, as shown in figure 3. As many as five different wave groups or phases can emerge when a P or S wave encounters a discontinuity or interface within the Earth.

Figure 3. Propagation paths of combinations of P, S, and L waves from an earthquake focus.

SOURCE: U.S. Geological Survey