When drops of rain fall, some of them flow downhill, each carrying a bit of dirt. As a result, over time the landscape changes from steep to shallow slopes. The land surface erodes away toward sea level. Like a boat in the water, Earth's continental crust displaces its own weight in the underlying layer of denser mantle rock. As its cargo of eroding soil and rock is thrown overboard, the crust loses weight and rises. To accommodate this change, some hot mantle rock must flow in beneath the continent, just as water flows in under a boat that is rising out of the water as its load is lightened. The only difference is the speed at which it happens: hot rock flows slowly, a finger's width a year.
The mantle rock that flows inward under the thinning continent must come from somewhere and, in turn, be replaced by other hot mantle rock. Where does it come from? We know that fresh mantle rock rises up at the mid-ocean ridges, where tectonic plates pull apart to release hot material coming up from below (Smithsonian, January and February 1975). Some of that mantle rock becomes oceanic crust, adding to the plates' trailing edges as they move apart. And some of it flows underneath the oceanic crust to fill the space being created by that lightening, rising continental crust. All this moving mantle rock is, in turn, replaced by the upward flow of hot rock from deeper in the planet. We face the astonishing fact that raindrops falling on land indirectly cause hot, flowing rock material to rise up from Earth's depths.
Until it nears the surface, the flowing rock we have seen in action is flowing in a plastic way, deforming like steel squeezed between rollers. But when hot rock rises from the interior under ocean ridges, it partly melts into a liquid as the pressure on it lessens. The melted rock travels through pores and cracks until it collects and rises and erupts in submarine volcanoes. The cooling lava, transferring its heat to the water, helps the Sun heat the ocean, powering the wind and producing the rain.
One drop of water; mantle flow; volcanism; rain. Loop closed. We are back at the beginning of the cycle.
Eventually, mantle rock flowing beneath the oceanic crust encounters, pushes and slides past the colder, more-rigid rock material of the continent. Something in that crust breaks, and there is an earthquake, because cold rocks break rather than flow. One drop of water. One Northridge earthquake.
The flowing mantle eventually cools and dives back down into the earth in subduction zones, producing more earthquakes and volcanoes as it does, especially in the "ring of fire" around the Pacific. No ocean floor is older than 180 million years; all of it goes down the tubes sooner or later.
If a piece of tectonic plate is subducted, its place at the surface is taken by new material somewhere else, at an ocean ridge. In North America we ride westward away from the Atlantic Ridge on the North Atlantic Plate; at the Ridge new rock is made from magmas that come up from hot mantle. Meanwhile, the western edge of the Pacific Plate slides down into the mantle. To paraphrase John Donne, no plate is an island, entire of itself. All motions are interrelated. One drop of water may set the whole thing going.
Our planet knows only one thing: how to get rid of heat. The silicate mantle, amounting to two-thirds of Earth's mass, contains traces of radioactive elements that heat the planet a bit as they decay. Beneath the mantle is the hot metallic core, most of the other third of Earth. The cooler part of the descending mantle eventually touches the core. Heat leaks upward out of the molten core faster than usual through this cold finger. Such cooling causes a little bit of the liquid-metal outer core to sink and solidify as crystals onto the solid-metal inner core at the center of the planet.
Light, nonmetallic elements like sulfur are released at this boundary. This light material rises, stirring the liquid outer core in the process. Responding to Earth's rotation, it swirls in a spiral up toward the rocky mantle. Flowing electrons in the molten nickel and iron produce an electric current. Earth rotates the electric current, making a dynamo and thereby producing a magnetic field. Cold rock replacing hot rock at the core causes a little of the liquid outer core to solidify, setting off a train of events that result in a magnetic field. Raindrops are the cause of Earth's magnetic field? Maybe just a little bit.
If a disturbance in the mantle interferes with the flowing source of the magnetic field, it may die away and reverse its polarity (the magnetic north pole will become the south pole, and vice versa). Such reversals happen every few hundred thousand years or so. When Earth's magnetic field is weak during the switch, charged particles that are normally trapped in that field hit the surface all over the globe instead of only at the magnetic poles (where they cause the auroras borealis and australis). This ionizing radiation damages the DNA of all living things. Rain causes mutations? We are getting far from the cause here, but the thread still holds.
Some of the atmosphere's moisture falls to earth as snow. In some cold places this stays year-round, making ice sheets and their peripheral glaciers. The weight of the ice sheets presses down on the rocks underneath, and the base of the continent sinks into the mantle. The mantle flows away from that place. If the ice sheet becomes very thick, the heat rising into it from within Earth (doing what Earth does best) may melt the bottom layer of the ice. When this happens for long enough, the ice rides away on a film of water and crushed rock at a rate of perhaps a finger's width a minute. It reaches the ocean and breaks up into giant icebergs. These sail across the North Atlantic to Spain and France, changing the circulation of the ocean as they go. This, in turn, could cause the ice to begin to grow again, or maybe not, depending on the trend of things that we still do not understand (Smithsonian, August 1993 and September 1995). If the ice sheet then melts away, the land bounces ! back up and the rocky mantle flows right back into place beneath it. One drop of water (disguised as a snowflake); one ice sheet gone; one rebounding continent exposed.
When, you ask, did this ever happen? Oh, only about 10,000 years ago and many times before that for 70,000 years (and, we expect, for thousands of years into the future); only a tick in time for Earth. The record of these glacial splurges lies in the sediment at the bottom of the ocean.
Some of the drops of water that fall to the ground don't just flow to the sea: we drink them first. Some have been in the ground a long time (the water we drink today may have been in the ground for a thousand years). We are glad of fresh water but may use it up without suspecting we are mining a resource. A millennium of tree ring records in the rain shadow of the Sierra Nevada show that what they call drought in Reno is really normal. The export of Colorado River water westward was premised on high flows that have existed only rarely in the long term.
There is just one Earth. It is a closed system except at the outer reaches of the atmosphere; except for meteorite infall; and except for the blast of heat and subatomic particles from the Sun. (And except for invisible but very real gravitational fields: the Sun and Moon cause the tides that clean our shores twice a day.)
The point of this essay is to demonstrate the unity of Earth processes and therefore the unity of Earth research. If one drop of water affects the whole Earth, all research bearing on our planet is applied research. The distinction between pure and applied research in earth science is fuzzy and artificial. Earth's inevitable, enduring pulse will govern our activities whether we recognize it or not. We should pay attention. If we are arrogant about Earth, it will make us poor. If we understand Earth, it will make us rich.
One drop of water. Knowing it can help us prosper.
By Stearns A. Morse
Stearns A. Morse, professor of geology at the University of Massachusetts, Amherst, marks the 75th anniversary of the American Geophysical Union by including the work of all ten of its sections in this column.