Infer the composition and properties of Earth's interior from the behavior of seismic waves

Learning goal: Explain how the behavior of seismic waves, together with density and magnetism evidence, identifies the composition and properties of Earth's interior; describe the crust, mantle and core; and account for where Earth's internal thermal energy comes from and what it drives.

Nobody has been there. The center of Earth is about 6,371 km down, far deeper than any drill hole or mine has ever reached. Almost everything known about the interior is inferred from evidence gathered at the surface. The most important evidence is the behavior of seismic waves: the vibrations released by earthquakes (and by large explosions) that travel through the whole planet and are recorded by instruments called seismometers on every continent. A wave's speed and path depend on the properties of the material it passes through, so waves that have traveled through the interior carry information about it back to the surface.

Two kinds of body waves. Seismic waves that travel through Earth's body come in two types, and the difference between them is the key to the whole argument. A P-wave (primary wave) is a compressional wave: the material is pushed and pulled back and forth along the direction the wave travels, like a pulse sent along a stretched spring. P-waves can travel through solids, liquids and gases, because every kind of matter resists being squeezed. An S-wave (secondary wave) is a shear wave: the material moves sideways, at right angles to the direction of travel, like a rope shaken up and down. Shear waves need a material that resists being bent out of shape. Solids resist that; liquids and gases do not. S-waves therefore cannot travel through a fluid, whether liquid or gas. P-waves are also faster than S-waves in the same material (roughly 6 km/s against about 3.5 km/s in crustal rock, though both values change with the rock), which is why the P-wave is the first vibration a distant seismometer records. At this grade, reading the arrival times off a seismogram is not the goal; what matters is that the two wave types have different properties, and those properties reveal the interior.

Refraction and reflection. When a wave passes from one material into another with different properties, its speed changes, and if it crosses the boundary at an angle the change in speed bends the wave's path (a wave that meets the boundary head-on changes speed without bending). That bending is refraction. Part of the wave's energy also bounces back from the boundary; that is reflection. Seismologists use both. Waves reflected from a boundary tell them the boundary exists and how deep it is. Waves refracted across a boundary tell them the material on the other side has a different speed, and therefore a different density, stiffness or state. Where wave speed increases steadily with depth, paths curve gradually upward and return to the surface; where speed changes suddenly, paths kink. Reading the pattern of arrivals at many stations is how the interior is mapped, layer by layer.

The shadow zones. After a large earthquake, seismometers around the world record the waves, but not everywhere. On the far side of the planet from the earthquake, beyond an angular distance of about 103 to 104 degrees measured from the earthquake, no direct S-waves arrive at all. Something deep inside stops them. Since S-waves cannot pass through liquid, the explanation is that part of the deep interior is liquid: the outer core. P-waves also show a gap, a ring-shaped P-wave shadow zone from about 104 to about 140 degrees away, where direct P-waves are missing. P-waves do get through the liquid, but they slow down sharply when they enter it (in round numbers, from an average of about 13 km/s in the lower mantle to an average of about 9 km/s in the outer core), so they refract strongly and are bent away from that ring. The shadow zones are not gaps in the instruments; they are the fingerprint of a liquid layer with a definite size. From the width of the zones, the radius of the core has been worked out to about 3,500 km. Waves that pass through the very center behave as if they meet a solid again, which is the evidence for a solid inner core with a radius of about 1,220 km.

Density evidence. Earth's mass and volume are known from gravity and its size, and dividing them gives an average density of about 5.5 g/cm³. Crustal rock averages only about 2.8 g/cm³. An average is weighted by volume, so if the outside is lighter than the average, the inside must be heavier than the average. On its own that is all the comparison shows: the crust is such a thin skin, about 1 percent of Earth's volume, that the rest of the planet need only average a little above 5.5 g/cm³. It is the seismic evidence that says where the dense material is: the mantle, most of Earth's volume, transmits waves at speeds that match rock of roughly 3.3 g/cm³ near the top to about 5 g/cm³ in the lower mantle, so for the planet to average 5.5 the core must be far denser still. Combined, the evidence points to a core made mostly of iron with some nickel, with a density of roughly 11 g/cm³ in the outer core to about 13.5 g/cm³ in the inner core. Iron meteorites, thought to be fragments of the cores of small bodies broken up early in the solar system's history, are made of iron and nickel, which supports the composition claim from a second direction. Why the iron is at the center is explained by planetary differentiation: while the young Earth was hot enough to be largely molten, the densest material, mostly iron and nickel, sank toward the center and lighter material stayed nearer the surface, producing layers with different composition.

Magnetism evidence. Earth has a magnetic field, but it is not a giant permanent bar magnet. Iron loses its permanent magnetism above about 770 °C, and the core is several thousand degrees. The field must be generated continuously by motion: flowing, electrically conducting liquid iron in the outer core acts as a dynamo. That the field exists at all is therefore further evidence that the outer core is liquid metal, in motion.

The layers. Putting the evidence together: the crust is the thin solid outer skin, roughly 5 to 10 km thick under the oceans and 20 to 70 km under continents. The mantle is solid rock extending to about 2,900 km depth; it is the largest layer by volume. The outer core is liquid iron and nickel, from about 2,900 km down to about 5,150 km. The inner core is solid iron and nickel, squeezed solid by enormous pressure even though it is the hottest layer. The diagram draws these four layers with thickness in proportion, except that the crust is drawn thicker than true scale so that it is visible at all. Two ways of dividing the interior are in use. By composition, the layers are crust, mantle and core. By physical behavior, the same material is divided into the rigid lithosphere (the crust plus the uppermost mantle), the weak asthenosphere beneath it, the stiffer lower mantle (the mesosphere) and the outer and inner core. The boundaries are also less sharp than a diagram suggests: each layer's thickness varies from place to place, and some transitions are gradual, depending on composition, density, temperature and pressure.

Where the heat comes from. Two sources supply Earth's internal thermal energy. First, during planetary formation, countless impacts of colliding bodies converted gravitational potential energy into heat, and a great deal of that original heat is still stored deep inside, because rock conducts heat very slowly. Second, the decay of certain unstable atoms inside Earth, in the crust and mantle and possibly in the core as well, continuously releases thermal energy; together with the stored heat, this is why the interior has not simply cooled off in 4.6 billion years. (How atoms decay is a topic for later courses; here it is enough to know that decaying atoms are an ongoing heat source.) A third, smaller contribution comes from friction as material moves and shifts inside Earth. Thermal energy moving outward from the core heats the lower mantle, and this energy drives convection currents in the asthenosphere, the hot, weak part of the upper mantle beneath the rigid plates. A common misconception is that the asthenosphere is an ocean of liquid magma. It is not: seismic S-waves pass through it, so it is solid. It is solid rock hot enough to deform and creep very slowly, a few centimeters a year, and that slow convection is what the next lesson connects to moving plates.

What the evidence cannot tell us. Seismic waves give the physical properties of each layer: state, density, stiffness and the depth of boundaries. They do not directly give chemical composition; that is inferred by combining density with what is known about meteorites and about how minerals behave under pressure. A careful explanation says which claims come from which evidence.

Worked density argument: A student measures a piece of granite: mass 270 g, volume 100 cm³ (found by water displacement). Density = 270 ÷ 100 = 2.7 g/cm³. Earth's average density is about 5.5 g/cm³. Since the average is weighted by volume, and the outermost rock is only 2.7 g/cm³, material somewhere inside must be above 5.5 g/cm³ to bring the average up. A denser-than-rock interior is required; the granite alone cannot say how deep, how dense or how much, only that surface rock is not representative of the whole planet. Because the crust is so thin, the comparison by itself does not even show that the interior is much denser than 5.5; that comes from the seismic evidence about the mantle.

Worked shadow-zone reasoning: A large earthquake occurs. Stations up to about 103 degrees away record both P- and S-waves. Stations beyond that record P-waves (outside the P shadow ring) but never direct S-waves. Claim: part of the interior is liquid. Reasoning: S-waves are shear waves and shear waves cannot travel through a fluid (liquid or gas); the waves that would have reached the far side had to pass through the deep interior; they did not arrive; therefore the deep interior they would have crossed is liquid. Condition: this identifies the outer core as liquid. It does not show that the mantle is liquid, because S-waves that pass only through the mantle do arrive.

Worked refraction reasoning: A P-wave traveling down through the mantle speeds up with depth, so its path curves gently. At the core boundary its speed drops suddenly: in round numbers, from an average of about 13 km/s in the lower mantle to about 9 km/s in the outer core. A sudden speed drop bends the path sharply toward the center. Waves that were headed for the region 104 to 140 degrees away are bent past it, which is why that ring receives no direct P-wave. So the shadow zone is a consequence of refraction at a boundary, not of the wave being absorbed.

Worked energy account: Question: if Earth formed 4.6 billion years ago, why is the interior still hot? Answer in two parts. (1) Formation impacts converted gravitational potential energy to heat, and rock is such a poor conductor that much of that original heat remains. (2) Decaying atoms inside Earth release thermal energy continuously, adding new heat every year. Together these keep the core hot enough to drive convection in the asthenosphere.

Representation task: Draw a cross-section of Earth to the same proportions as the lesson diagram (crust, mantle, outer core, inner core) and label the approximate depth of each boundary. Then draw two wave paths from one earthquake: an S-wave that reaches a station 90 degrees away, and an S-wave aimed at a station 150 degrees away. Show where the second path stops and write one sentence beside it saying why.

Safe investigation, wave types and density: (a) Lay a long spring toy on the floor with a partner holding the far end. Send a push-pull pulse along it and sketch how the coils move relative to the direction of travel: this models a P-wave. Send a side-to-side pulse and sketch it: this models an S-wave. Time each kind of pulse over the length of the spring and repeat five times. (b) Measure the mass of three different rocks on a balance and their volumes by water displacement in a measuring cylinder, and calculate each density. Compare with Earth's average of about 5.5 g/cm³. What the model cannot show: a spring is not rock; in a stretched spring toy the two kinds of pulse usually travel at about the same speed (your timings will show whether yours does), so the model does not reproduce the real speed difference between P- and S-waves; the spring does not let you see a shear pulse fail in a liquid; and three surface rocks cannot reveal how density changes with depth; those claims still rest on the seismic evidence.

Independent performance: Write an explanation of at least 250 words titled "How we know what is inside Earth without going there." Success criteria: it names both body wave types and states how the two differ, including the key difference in what they can pass through (S-waves cannot pass through a liquid); it explains refraction and reflection at a boundary; it uses the shadow zone to identify the liquid outer core; it uses average density to argue for a dense interior; it names both sources of internal thermal energy; and it states one thing seismic waves do not directly tell us.