Seismic Study of Earth’s Interior

Seismic Study of Earth’s Interior 

Let’s now utilize your knowledge of seismic velocity, refraction, and reflection to see how each of the major layer boundaries inside the Earth was discovered.

Discovering the Crust-Mantle Boundary 

Discovery of the Moho.

The concept that seismic waves refract at boundaries between different layers led to the first documentation of the core-mantle boundary. In 1909, Andrija Mohorovicic, a Croatian seismologist, noted that P-waves arriving at seismometer stations less than 200 km from the epicentre travelled at an average speed of 6 km per second, whereas P-waves arriving at seismometers more than 200 km from the epicentre travelled at an average speed of 8 km per second. To explain this observation, he suggested that P-waves reaching nearby seismometers followed a shallow path through the crust, in which they travelled relatively slowly, whereas P-waves reaching distant seismometers followed a deeper path through the mantle, in which they travelled relatively rapidly (figure above a, b).

Calculations based on this observation require the crust mantle boundary beneath most continental regions to be at a depth of about 35 to 40 km. Later studies showed that the crust-mantle boundary beneath oceanic regions lies at a depth of about 7 to 10 km. The crust-mantle boundary is now called the Moho, in honour of Mohorovicic.

Defining the Structure of the Mantle 

The velocity of P-waves in the mantle changes because the physical properties of the mantle change with depth.

By studying travel times, seismologists have determined that seismic waves travel at different speeds at different depths in the mantle. Between a depth of about 100 and 200 km in the asthenosphere beneath oceanic lithosphere, seismic velocities are slower than in the overlying lithospheric mantle (figure above a). This 100- to 200-km-deep layer is called the low-velocity zone (LVZ); here, the prevailing temperature and pressure conditions cause peridotite to partially melt by up to 2%. The melt, a liquid, coats solid grains and fills voids between grains. Because seismic waves travel more slowly through liquids than through solids, the coatings of melt slow seismic waves down. In the context of plate tectonics theory, the low-velocity zone is the weak layer on which oceanic lithosphere plates move. Below the low-velocity zone, the asthenosphere does not contain melt. (Of note, seismologists do not find a well-developed low-velocity zone beneath continents.) 

Below about 200 km, seismic-wave velocities in the mantle increase with depth (figure above b). Seismologists interpret this increase to mean that mantle peridotite becomes progressively less compressible, more rigid, and denser with depth. This proposal makes sense, considering that the weight of overlying rock increases with depth, and as pressure increases, the atoms making up rock squeeze together more tightly and are not so free to move. Because of refraction, the increase in seismic  velocity with depth causes seismic rays to curve in the mantle. To understand the shape of a curved ray, let’s represent a portion of the mantle by a series of imaginary layers, each of which has a slightly greater seismic-wave velocity than the layer above (figure above c). Every time a seismic ray crosses the boundary between adjacent layers, it refracts a little toward the boundary. After the ray has crossed several layers, it has bent so much that it begins to head back up toward the top of the stack. Now if we replace the stack of distinct layers with a single layer in which velocity increases with depth at a constant rate, the wave follows a smoothly curving path (figure above d). 

At depths between 410 km and 660 km, seismic velocity increases in a series of abrupt steps (figure above a), so the stack of layers in figure above c is actually a somewhat realistic image. A major step occurs at a depth of 660 km. Experiments suggest that such seismic-velocity discontinuities occur at depths where pressure abruptly causes atoms in minerals to rearrange and pack together more tightly, thereby changing the rock’s compressibility and rigidity. Researchers are still unsure if chemical changes also occur at the discontinuities. Because of these seismic-velocity discontinuities, now can see more clearly, seismologists subdivide the mantle into the upper mantle (above 660 km), the transition zone (between 410 and 660 km), and lower mantle (below 660 km). Note that seismologists consider the transition zone to be part of the upper mantle. 

Discovering the Core-Mantle Boundary 

Shadow zones and the discovery of the Earth's core. 

During the first decade of the 20th century, seismologists installed seismometers at many stations around the world, expecting to be able to record waves produced by a large earthquake anywhere on Earth. In 1914, one of these seismologists, Beno Gutenberg, discovered that P-waves from a given earthquake do not arrive at seismometers lying in a band at a distance of between 103° and 143°, as measured along the surface of the Earth from the earthquake epicentre. This band is now called the P-wave shadow zone (figure above a). If the density of the Earth increased gradually with depth all the way to the centre, the shadow zone would not exist because rays passing into the interior would curve up and reach every point on the surface. Thus, the presence of a shadow zone means that deep in the Earth a major interface exists where seismic waves abruptly refract down (implying that the velocity of seismic waves suddenly decreases). This interface, the core-mantle boundary, lies at a depth of about 2,900 km. 

To see why the P-wave shadow zone exists, follow the two seismic rays labelled A and B in figure above a. Ray A curves smoothly in the mantle (we are ignoring seismic-velocity discontinuities in the mantle) and passes just above the core mantle boundary before returning to the surface. It reaches the surface 103° from the epicentre. In contrast, Ray B penetrates the boundary and refracts down into the core. Ray B then curves through the core and refracts again when it crosses back into the mantle. As a consequence, Ray B intersects the surface at more than 143° from the epicentre.

Discovering the Nature of the Core 

Based on the study of meteorites thought to be fragments of a large planetesimal’s interior, and on density calculations, seismologists concluded that the core consists of iron alloy. This alloy contains about 85% iron, 5% nickel, and 10% of a lighter element (probably oxygen, silicon, and/or sulphur). But the downward bending of seismic waves when they pass from the mantle down into the core indicates that seismic velocities in, at least, the outer core are slower than in the mantle. Thus, even though the core is deeper and denser than the mantle, at least the outer part of the core must be less rigid than the mantle. How can this be?

Seismologists found that S-waves do not arrive at stations located between 103° and 180° from the epicentre (a band called the S-wave shadow zone). This means that S-waves cannot pass through the core at all otherwise, an S-wave headed straight down through the Earth would appear on the other side. S-waves are shear waves, which by their nature can travel only through solids. Thus, the fact that S-waves do not pass through the core means that the core, or at least part of it, consists of liquid (figure above b). 

At first, seismologists thought that the entire core might be liquid iron alloy. But in 1936, a Danish seismologist, Inge Lehmann, discovered that P-waves passing through the core reflected off a boundary within the core. She proposed that the core is made up of two parts: an outer core consisting of liquid iron alloy and an inner core consisting of solid iron alloy. Lehmann’s work defined the existence of the inner core but could not locate the depth at which the inner core-outer core interface occurs. This depth was eventually located by measuring the exact time it took for seismic waves to penetrate the Earth, bounce off the inner core–outer core boundary, and return to the surface (figure above c). The measurements showed that the inner core–outer core boundary occurs at a depth of about 5,155 km. 

Why does the core have two layers a liquid outer layer and a solid inner one? An examination of figure above d provides some insight. This graph shows two curves: (1) the geotherm, which indicates how temperature changes with increasing depth in the Earth; and (2) the melting curve, which indicates that the temperature at which materials melt changes with increasing depth in the Earth. At temperatures to the right of the melting curve, Earth materials are molten. As the graph shows, the geotherm lies to the left of the melting curve through most of the mantle and in the inner core. This means that temperatures in most of the mantle and in the inner core are not high enough to cause melting, under the very high pressures found in these regions, so these regions are solid. But the geotherm lies to the right of the melting curve in the low-velocity zone of the mantle and in the outer core, so these regions contain molten material. 

A Modern Image of Earth’s Layers 

The velocity-versus-depth profile of the whole Earth.

Through painstaking effort, seismologists used data from this array to develop a graph, known as a velocity-versus depth curve, that shows the depths at which seismic velocity suddenly changes. These changes define the principal layers and sub-layers in the Earth (figure above). Note that the graph does not show a velocity for S-waves in the outer core because S-waves cannot travel through molten iron (a liquid). 

Seismic Tomography 

In recent years, seismologists have developed a technique, called seismic tomography, to produce three-dimensional images of variation in seismic velocities in the Earth’s interior. This technique resembles the method used to produce three-dimensional CAT (or CT) scans of the human body. In seismic tomography studies, researchers compare the observed travel time of seismic waves following a specific ray path with the predicted travel time that waves following the same path would have if the average velocity-versus-depth model depicted by Figures 2 and and 3 were completely correct. 

Tomographic images of the Earth’s interior, and their interpretation.

Tomographic studies emphasize that the simple onion-like layered image of the Earth, with velocities increasing with depth at the same rate everywhere, is an oversimplification. In reality, the velocities of seismic waves vary significantly with location at a given depth. Results of tomographic studies can be displayed by three-dimensional models, cross sections, or maps (figure above a,  b). Generally, warmer colours (reds) on these images indicate slower and presumably warmer regions, whereas cooler colours (blues and purples) indicate faster and presumably cooler regions.

Even though many important questions remain, tomography has led geologists to picture the Earth’s insides as a dynamic place (figure above c). This image should become even clearer, because a major research initiative, called Earth Scope, has begun. This initiative involves placing hundreds of seismometers in an array across the United States. Just as digital photographs have higher resolution when taken by a camera with a 14-mega-pixel sensor than one with a 2-mega-pixel sensor, the greater number of seismometers in the Earth Scope array provides a higher-resolution tomographic image of the Earth’s interior. 

Seismic-Reflection Profiling 

Seismic reflection profiling. The method allows geologists to see underground.

Seismic techniques are also letting us fine-tune our image of the upper crust. During the past half century, geologists have found that by exploding dynamite, by banging large weights against the Earth’s surface, or by releasing bursts of compressed air into the water, they can create artificial seismic waves that propagate down into the Earth and reflect off the boundaries between different layers of rock in the crust. By recording the time at which these reflected waves return to the surface, geologists can determine the depth to these boundaries. With this information, they can produce a cross-sectional view of the crust called a seismic-reflection profile (figure above a, b). This image can define subsurface bedding and stratigraphic formation contacts, and can reveal the presence of subsurface folds (bends in layers) and faults. Oil companies obtain many seismic-reflection profiles, despite their high cost, because they allow geologists to identify likely locations for oil and gas reserves underground.

In the past decade, computers have become so sophisticated that geologists can now produce three-dimensional seismic-reflection images of the crust. From the 3-D data, they can produce profiles and maps in any orientation desired. These provide so much detail that geologists can even trace out a ribbon of sand, representing the channel of an ancient stream, now buried kilometres below the Earth’s surface.

Credits: Stephen Marshak (Essentials of Geology)