Why Is the Sea Floor So Young?

You met the age of the Earth in Grade 8 Grade 8: Origin and age of the Earth: about 4,600 million years (4.54 billion, more exactly), measured from radioactive atoms in rocks and meteorites, which change into other atoms at a steady rate, like sand running through an hourglass. Measure the rocks of the Earth’s surface the same way and they split into two groups:

Where	The oldest found	Typical age
continents	rock: 4,030 million years (Acasta Gneiss, Canada); single grains of the mineral zircon: 4,400 million years (Jack Hills, Australia)	large areas over 2,500 million years
sea floor	about 180 million years (western Pacific)	average about 64 million years

Rocks this old can be measured only because the continents keep them. The sea floor covers about 60% of the Earth’s surface, yet none of it is old. Something must be making new sea floor and getting rid of old sea floor, but leaving the continents alone. To find out what, scientists had to look at the bottom of the sea, which nobody could see.

Look at a map of the Atlantic Ocean. The bulge of Brazil would fit into the curve of Africa’s west coast, like pieces of a jigsaw puzzle. Many people noticed this once good maps were made. One man made it into a theory.

The great southern land that India was part of is called Gondwana, after the Gond region of central India, where its rocks, coal and Glossopteris fossils were first studied. You will meet India’s journey north from Gondwana in Why Is Everest Still Rising? (India meets Asia).

Until the 1920s, the depth of the sea was measured by lowering a weighted rope until it touched the bottom: hours for one measurement in the deep ocean. Then ships began to use echo sounding: send a pulse of sound down, and time its echo from the sea floor. Sound travels through sea water at about 1,500 m/s.

By the 1950s ships had crossed the oceans with echo sounders running, and the record was turned into maps. The sea floor was nothing like the flat, muddy plain people had imagined.

In 1962 the American geologist Harry Hess put the clues together. Hot rock rises from the mantle under the ridges, melts, and erupts as lava, making new sea floor along the rift. The new rock is pulled away on both sides, making room for more. Far away, at the deep ocean trenches, old sea floor sinks back into the mantle. So the sea floor moves like a slow conveyor belt, and the continents ride on it rather than plowing through it. This is sea-floor spreading (a name given by Robert Dietz in 1961). Hess called his own paper “geopoetry”: a good story, but without proof. The proof was hidden in the sea floor’s magnetism.

You learned in Grade 8 that the Earth is a magnet, because of molten iron moving in its outer core, and that its field has flipped many times: the magnetic north and south poles swap places Grade 8: Why geomagnetism matters. A time when the field points as it does today is called normal; a time when it points the other way, reversed. A flip takes a few thousand years, and they come at irregular times, sometimes a few tens of thousands of years apart, sometimes tens of millions. The last one was 773,000 years ago: early humans living then would have seen their compasses, if they had had any, point south.

How can anyone know? Lava. Volcanic lava contains grains of the magnetic mineral magnetite. When it is hot, the grains are not magnets: heating destroys magnetism, as you saw in Grade 8 Grade 8: Heating the magnet. As the lava cools below a few hundred degrees Celsius (580 °C for pure magnetite; lower for the titanium-rich kind in sea-floor basalt), each grain becomes a tiny magnet lined up with the Earth’s field at that moment, and stays that way when the rock is solid. Every lava flow is a fossil compass. In the 1960s geologists measured the magnetism of lava flows all over the world and dated them, and built a timetable of the field’s flips: the magnetic reversal time scale.

In the 1950s, ships began towing magnetometers, instruments that measure the strength of the magnetic field, across the oceans. Over the sea floor the field was a little stronger in some places and a little weaker in others, and when the measurements were put on a map, the strong and weak places made long stripes, parallel to the mid-ocean ridges. Nobody knew why.

In 1963 the British geologists Fred Vine and Drummond Matthews explained them, using Hess’s idea. If new sea floor is made at the ridge and carried away, then each band of new rock records the field at the time it cooled. Rock made while the field was normal adds to today’s field and makes it a little stronger; rock made while it was reversed makes it a little weaker. Each flip leaves a stripe, and the spreading carries half of it each way. So the stripes on one side of a ridge should be a mirror image of the stripes on the other, and their widths should follow the timetable of the flips.

That is exactly what the ships found. In 1966 a profile across the Pacific–Antarctic Ridge, measured by the research ship Eltanin, was so perfectly symmetric that it could be folded down the middle and the two halves would match; and its pattern matched the reversal timetable from lava on land, stretched by the spreading speed. Sea-floor spreading had its proof.

The stripes do more than prove spreading: they measure its speed. Divide the width of a stripe by the time it took to make.

The magnetic stripes gave ages by matching patterns. Could anyone check them by bringing up the sea floor itself? In 1968 a new ship, the Glomar Challenger, began the Deep Sea Drilling Project. It could hold its place over water 6 km deep while it lowered a drill pipe, kilometers long, to the sea floor and drilled into it.

Mud settles on the sea floor all the time: dust blown from land, and the tiny shells of plankton that sink when they die. It can only settle on sea floor that already exists. So the oldest mud lying on the hard basalt rock tells you, roughly, when the basalt was made. Tiny fossils in the mud give its age, because different kinds of plankton lived at different times.

On its third voyage, from December 1968 to January 1969, the Glomar Challenger drilled a line of holes across the Mid-Atlantic Ridge in the South Atlantic, near 30° S. Here are the real results.

The ages rose steadily with distance from the ridge, on both sides, just as the magnetic stripes said they would, and gave the same speed: about 2 cm a year on each side. Since then, ships have drilled thousands of holes in every ocean. No sea floor has been found older than the stripes predict.

The sea floor is not made only to stay. As a piece of it moves away from the ridge, it cools. Cool rock is denser, so the old sea floor sinks lower (that is why the sea is deeper far from the ridges, as in the echo-sounding try above), and the cold rock beneath it thickens the plate. After a few tens of millions of years, the plate of old sea floor is denser than the hot mantle under it. Where it meets another plate, it bends down and sinks into the mantle, making a deep trench. This sinking is called subduction.

So every piece of sea floor has a life: born at a ridge, carried across an ocean, and sunk at a trench, usually within 200 million years. Here is the whole sea floor, colored by age.

Because continents can’t sink, they keep almost all the rock they have ever had. And they grow. At a trench, volcanoes add new rock to the continent’s edge; islands, slivers of sea floor and small pieces of continent carried in on the sinking plate get scraped off and stuck on; and when two continents collide, their edges are crushed together into mountains. So a continent is usually oldest in the middle, at its core, and younger towards its edges, where pieces have been added on.

North America shows it clearly. Its core, around Hudson Bay, is rock over 2,500 million years old. Around it are belts 1,800–1,600 million years old, then belts 1,400–1,000 million years old, from collisions that welded new pieces on, then the Appalachian Mountains in the east, made in collisions about 300–450 million years ago, and the young mountains of the west, still being added to today.

India has old cores too: rocks over 3,000 million years old in the south and east of the country. The Himalaya are the reverse of what you might expect: very young mountains, still rising, made largely of old rock from India’s northern edge, some of it more than a billion years old, with the younger mud laid down in the Tethys Sea along India’s northern edge, shaligrams and all, piled on top.

Stripes and drill holes show how the plates moved over millions of years. Are they moving now? Since the 1990s, GPS receivers fixed to rock all over the world have measured it directly. Like the GPS in a phone, but far more precise, they measure their position to within a few millimeters, and they all move, year after year, by a few centimeters, in the directions the stripes say. Iceland, which sits on the Mid-Atlantic Ridge, is being pulled apart by about 2 cm a year; Hawaii moves north-west by about 7 cm a year; and stations in Nepal show India still pushing into Asia (see Why Is Everest Still Rising?, Where plates meet).

There is another record of the past in the middle of the Pacific Plate. A hot plume of rock rises from deep in the mantle under Hawaii and makes volcanoes. The plate slides over it, carrying each volcano away, so the islands form a line, oldest farthest away.

• Why Is Everest Still Rising? The heat inside the Earth that drives the plates, how we know what the inside is like, and what happens where plates meet, as India pushes into Asia.
• How Old Is the Earth? How the radioactive clocks in rocks, meteorites and tiny zircon crystals give their ages, and what the oldest rocks tell us about the young Earth.
• How Do Magnets Push Without Touching? Magnetic fields, the invisible pattern the sea floor records.