North Pole, South Pole: The Epic Quest to Solve the Great Mystery of Earth's Magnetism

by Gillian Turner Phd

  Through his friendship with Adam Sedgwick, a Fellow of Trinity College and inaugural professor of geology, Hopkins also developed a passion for geology, in particular the application of mathematics and physics to investigations of the Earth’s interior. Although often working on shaky assumptions (the reason his work is virtually ignored today), he would lay the groundwork for much of modern geophysics.

  It was clear that in the Earth’s interior there had to be layers of fluid whose temperature exceeded the melting point of rock-forming materials. Hopkins pointed out that with increasing depth a trade-off would happen: beyond a certain depth the increase in temperature would, other factors being equal, lead to melting— but the increase of pressure with depth would also raise the melting point, and if the pressure were high enough this might result in solidification. In an attempt to determine a “melting-point curve” for the Earth, he enlisted the help of his former student William Thomson, and another physicist, James Joule, who was by then famous for his experiments on heat and energy, to try and measure the melting points of rocks under high pressure. Unfortunately, these experiments proved rather too ambitious: the men reached no firm conclusions and Hopkins turned to other means of detecting fluidity within the Earth.

  It had long been known that Earth’s rotation axis shifted slightly but regularly in response to the gravitational pull of the sun and the moon. Astronomers had observed two separate effects, which they called precession and nutation. Hopkins figured that details of these tiny angular motions would differ depending on whether the Earth were rotating as a solid whole, or internal fluids were slipping and lagging behind the solid parts. His first set of mathematical calculations—nearly forty pages—was presented in the first part of his “Researches in Physical Geology” in the 1839 Philosophical Transactions of the Royal Society, but the calculations had proved inconclusive. Hopkins was still unable to distinguish between the models over which the geologists and physicists were arguing.

  However in 1842, in the third part of his “Researches,” Hopkins finally came down in favor of a predominantly solid planet, in which volcanic magma became truly fluid only with the release of pressure as it rose through near-surface fissures and cavities immediately prior to eruption. This must have influenced William Thomson, who subsequently declared the Earth to be “as rigid as steel,” and made his famously erroneous estimate of the age of the Earth as a few tens of millions of years on the basis that the planet could lose heat only by solid-state conduction.

  Many geologists and biologists vehemently disagreed, arguing that it must have taken far longer for the vast thicknesses of sedimentary rocks present on the Earth’s surface to have accumulated, or for life to have reached its present degree of evolution.

  New information on the shape and structure of the Earth would next come from detailed measurements of gravity at the surface, a study known as geodesy. Even before Isaac Newton’s publication of Principia in 1687, it had been noticed that a pendulum swung more slowly at the equator than at higher latitudes. In 1672 a French astronomer, Jean Richer, had noted that his pendulum-clock lost two-and-a-half minutes per day at the equator, compared with measurements made before he had left Paris. This meant that at the equator the force of gravity was slightly weaker and “g,” the acceleration due to gravity, slightly less than at Paris.

  Newton had explained this by suggesting that the planet bulged at the equator, so Richer’s pendulum-clock was further from the center of the Earth and therefore experienced a smaller gravitational force, causing it to swing more slowly. The equatorial bulge, Newton had argued, was a direct effect of Earth’s rotation. To make something revolve in a circle it must be pulled inwards (centripetal force): the smaller the circle or the faster the rotation, the bigger the pull that is needed. In order to provide this pull, Newton supposed that the Earth became stretched outwards at the equator.

  In the eighteenth century, the Swiss mathematician Leonhard Euler had predicted that, as a result of the equatorial bulge, Earth’s rotation axis should be displaced a few meters from the axis of symmetry, and should move around it. Like the precession and nutation studied by Hopkins, the detailed dynamics of Euler’s perturbation would depend on how mass was distributed within the Earth and how rigid it was—“as rigid as steel,” as Kelvin had proposed, or fluid, as many geologists inferred, or somewhere in between. If the Earth were rigid, calculations showed the period of Euler’s axial perturbation would be 305 days. However, when the effect was finally observed in 1891 by an American astronomer, Seth Carl Chandler, it was found to be 435 days, or about 14 months.

  The effect, now known as Chandler Wobble, meant that latitudes seemed to fluctuate very slightly in a fourteen-month cycle. This showed the planet could not be truly rigid: it must deform very slightly in response to the forces imposed on it. It was, in other words, “elastic.”

  Just as important as the physical state of Earth’s inner regions was the chemical composition of the materials that made it up. Surface rocks were one thing but what could there be deep down, either solid or liquid, that was so much denser than the rocks on the surface?

  In 1896 Emil Wiechert, a young geophysicist from Königsberg in Prussia, put forward a simple two-part model of the Earth that he claimed was compatible with all available astronomical and geodetic information. Observations of the precession and nutation of the axis of rotation had by now shown that the densest material must be concentrated very deep down, close to the center of the Earth. To account for this, Wiechert boldly proposed a core of completely different composition from the overlying rocky shell. Since the only materials known to have densities higher than Earth’s average value of 5500 kilograms per cubic meter were metals, he suggested that the core must be iron, the most abundant of the stable metallic elements. This made sense. Iron was known to be a common constituent of meteorites, and these were thought to be fragments of planetary bodies that had formed and broken up early in the history of the solar system.

  According to Wiechert’s calculations, the core had to have a density of 8200 kilograms per cubic meter and occupy nearly fifty percent of the volume of the planet, giving it a radius of some 5000 kilometers. The overlying shell had to have an average density of 3200 kilograms per cubic meter.

  Wiechert called this outer layer “der Mantel.” Some scientists objected on the grounds that a mantle was a loose, floppy cloak and so this was a poor description, but the word stuck. Meanwhile, Wiechert’s calculations would be repeatedly revised and improved. Today we know that the core’s radius is about half the total radius of Earth, and so the core occupies only about fifteen percent of Earth’s volume, with the mantle making up virtually all the remaining eighty-five percent.

  Amid the increasingly complex wobbles and perturbations of geodesy and the ever more detailed measurements of gravity, a new method of remotely sensing Earth’s interior had unexpectedly appeared on the scene. Ernst von Rebeur-Paschwitz, a German geodesist, had set up a delicate pendulum at Potsdam near Berlin, and another at the coastal town of Wilhelmshaven, 350 kilometers to the northwest. These pendulums were designed to measure the horizontal motion of the ground due to supposed lunar tides. According to Rebeur-Paschwitz, they would record this information automatically “by the same photographic method as that employed for magnetic observations.”

  On April 17, 1889, at very nearly but not quite the same time, both instruments were violently shaken by a sudden series of vibrations. Rebeur-Paschwitz was completely mystified until two months later he read in the June 13 issue of the scientific journal Nature of a great earthquake that had rocked Tokyo. He would later write:

  Reading the report on this earthquake, I was struck by its coincidence in time with a very singular perturbation registered by two delicate horizontal pendulums at the observatories of Potsdam and Wilhelmshaven.

  Taking local time differences into account, he calculated that the earthquake disturbances had taken an average of 3858 seconds— just over an hour—to travel fro
m Tokyo to Germany. As this was an average surface distance of 8264 kilometers, he estimated their speed to have been just over two kilometers per second. Considering this a reasonable figure for waves traveling close to the surface of the Earth, he concluded:

  … the disturbances noticed in Germany were really due to the volcanic [sic ] action which caused the earthquake of Tokio.

  Interestingly, this was not a new idea: it had been predicted at least half a century earlier that vibrational waves should propagate not only along Earth’s surface but also through its interior, and that their speed might shed light on the physical properties and composition of materials inside the Earth. French mathematicians Augustin-Louis Cauchy and Siméon-Denis Poisson had shown that energy should propagate through flexible solid materials in the form of waves, and William Hopkins had adapted the theory to the case of the Earth, and given a detailed exposition on the “vibratory motions of the Earth’s crust produced by subterranean forces” and the “observations required for the determination of the center of earthquake vibrations, and on the requisites of the instruments to be employed” to the 1847 meeting of the British Association for the Advancement of Science in Oxford.

  The physics behind this was relatively simple. When an elastic material is compressed, it responds by trying to expand back out. Conversely, when it is stretched (or “rarefied”) it tries to contract. If a material is alternately compressed and stretched, it responds by radiating out into the material around it a series of compressions and rarefactions: this motion is called a wave.

  Hopkins had predicted that two different types of elastic waves would travel through the body of the Earth. The first, which he called “normal vibrations,” would involve alternate compressions and rarefactions, with parts of the material moving forwards or backwards, parallel, or in the opposite direction to that in which the waves were traveling. Today these are known as P (“primary” or “pressure”) waves. They are identical to sound waves.

  Secondly, Hopkins said, there should also be “tangential vibrations” in which the material would undergo a shearing process, so parts of it moved at right angles to the direction in which the wave was traveling—rather like a wave on the surface of water. Today these are called S (“secondary” or “shear”) waves.

  Hopkins explained that in the same material a normal (primary) wave would always travel faster than a tangential (secondary) wave, since the speed depended on the compressibility of the material as well as its rigidity, or resistance to shearing motion. If an earthquake took place and the speeds in the Earth of the primary and secondary waves were known, it would be possible to work out, from the time between the waves’ arrivals at a certain spot, the distance to the quake’s epicenter. He envisaged doing this across a network of observation sites so earthquake centers could be accurately located by triangulation; he even described the requirements of the instruments that would be needed to measure the vibrations.

  Furthermore, whereas both types of waves could travel through solid materials, Hopkins pointed out that only primary waves could travel through fluids. A fluid, because it could flow, had no resistance to a shearing deformation and so could not sustain the sort of movement involved in the passage of secondary waves. Put simply, solids could transmit both P and S waves, with the P waves being faster, while fluids could sustain only P waves.

  Hopkins went on to predict that the study of earthquake disturbances would eventually enable scientists to reconstruct the paths of seismic waves through the different layers of Earth, and so decipher the planet’s hidden structure. Now, at last, RebeurPaschwitz had, albeit unintentionally, detected seismic waves, even if they were surface waves, rather than the body waves Hopkins had predicted. He had shown that with suitable instrumentation the waves’ travel times could be measured, and from this their speeds could be calculated. This information could then be interpreted to uncover the physical properties of the rocks through which the waves had traveled.

  Tragically, Rebeur-Paschwitz contracted tuberculosis and died in 1895 at the age of only thirty-four, before he had a chance to further develop his interest in seismology and the structure of the Earth. However, other scientists had by then recognized the need for a global network of earthquake observatories similar in concept to Gauss’s Göttingen Magnetische Verein. Seismology was up and running. Over the next few decades, observations, interpretations and new ideas about Earth’s interior would accumulate rapidly.

  In 1897, after a particularly violent earthquake in Assam, India, an English seismologist, Richard Dixon Oldham, finally succeeded in separating the arrival signals of P and S waves that had traveled through the interior of the Earth from those of the slower waves that had traveled along the surface. He suggested that the waves’ paths and speeds were consistent with the Earth having a glassy or stony mantle, and an iron core that extended from its center to just over half of its radius. Although this was remarkably close to modern estimates of the size of the core, Oldham made some assumptions that would come to be considered dubious and so today his calculation is rarely acknowledged.

  It would be Beno Gutenberg, Wiechert’s twenty-three-year-old student from Göttingen, who would come to be credited with the discovery of a sudden drop in seismic wave speed about halfway to the center of the Earth. The clue that led Gutenberg to propose such a “low-velocity” core was what has been called the “shadow zone.” Assuming that P and S waves radiated in all directions from the source of an earthquake, Gutenberg reasoned that waves traveling along shallow paths would reappear at the surface relatively close to the epicenter, while those traveling deeper down would reappear further away. However, the waves could reach only a certain distance, about 12,000 kilometers from the epicenter, without having to pass into the core.

  Gutenberg found waves that entered the core reappeared at Earth’s surface much further from the earthquake source than they would have done had they traveled through a completely uniform Earth. In fact, no P (or S) wave arrivals were detected anywhere between about 12,000 and 16,000 kilometers (105° and 143°) from an earthquake’s location. He inferred from this that when the waves entered the core they slowed, making their paths steepen, and when they re-emerged their paths were such that they could not reach the surface anywhere within the so-called “shadow zone.” From these observations, he calculated that the boundary between the mantle and the core must lie at a depth of about 2900 kilometers and that, on entering the core, seismic waves must suddenly slow to about 65 percent of their speed in the mantle.

  Gutenberg supposed both the mantle and the core to be solid. He envisaged a change in composition from a rocklike material with a P wave speed of about 12 kilometers per second to a possibly iron-rich core with a P wave speed of about ten kilometers per second. In this respect, his seismic model of the Earth was in agreement with Wiechert’s model, which had been based on density and on geodetic and astronomical observations. Each envisaged two parts, but the boundaries between the two parts lay at quite different depths in the two models.

  The question was, were the two scientists looking at the same transition, or were there two? By this time, other geophysicists had revived the idea of fluids deep inside the Earth, and were arguing that the transition (or transitions) corresponded to a change not from one solid state to another, but from a solid mantle to a liquid core. One strong argument for the core being liquid was the growing evidence that S waves did not seem to travel through it at all: unlike P waves, they did not reappear beyond the shadow zone. An even more compelling argument was that the average rigidity of the Earth, as deduced from tidal measurements of the sort studied by Rebeur-Paschwitz and from the Chandler Wobble, could simply not be reconciled with a completely solid Earth.

  The formation of a seismic shadow zone. Waves entering the core are slowed down and refracted so they reappear at the surface further from the source than they would have otherwise, leaving a shadow zone in which no waves are detected. In the Earth the shadow zone is observed between 105 a
nd 143 degrees from the source of an earthquake.

  It would be 1926 before a pronouncement by a noted British geophysicist would finally settle the argument. Harold Jeffreys, the author of the famous and much reprinted textbook The Earth, had improved Wiechert’s calculations and eventually come to the conclusion that Wiechert’s discontinuity in density and Gutenberg’s seismic discontinuity were one and the same. Furthermore, the discontinuity corresponded to changes both in composition (from rocky to iron-rich material) and in phase (from solid to liquid). He therefore declared that Earth’s core was “truly fluid.”

  Shadow zone of the magnitude 7.8 earthquake that struck Murchison, New Zealand on June 17, 1929. New Zealand lies out of sight on the bottom right reverse side of the globe. Cast by Earth’s core, the shadow zone, within which no P or S waves were expected, covers the area between 105° and 143° from the epicenter. Baku (B), Sverdlovsk (S) and Irkutsk (I), all lay within it.

  It seemed the Earth’s interior was at last understood. Geodesists, astronomers, seismologists, geologists and evolutionists were all in agreement. Even physicists, having discovered in radioactivity both another heat source within the Earth and a new means of dating rocks, revised their estimate of the age of the Earth to several thousand million years and joined the party.

  Seismograms from Baku, Sverdlovsk and Irkutsk all show unexpected P wave arrivals (indicated by arrows) from the New Zealand earthquake of June 17, 1929, a finding which led to the discovery of Earth’s inner core. The weak P waves are followed by much bigger surface waves.