Plate tectonics: the hidden key to life on Earth

“Look again at that dot. That’s here. That’s home. That’s us.” Carl Sagan was moved to lyricism by the pale blue dot that Voyager 1 photographed as it exited the solar system 27 years ago. The pale blue dot is precious, and lucky.

Not only does Earth lie in the ‘Goldilocks zone’ that allows water to exist in the liquid form that life requires. It is also the only rocky planet we know of that constantly renovates its surface as its tectonic plates dive into the mantle in some places and re-emerge as molten lava in others. Many astrobiologists now think this constant renewal is just as important as liquid water for the flourishing of life as we know it. 

The theory, explains planetary scientist Adrian Lenardic of Rice University in Houston, Texas, is that the Earth’s climate has been buffered by the recycling of carbon dioxide (CO₂) from the atmosphere into the planet’s interior via mineral sequestration and then out again via volcanoes. This has kept the climate temperate even as the Sun’s heat has increased in intensity by about a third since the planet’s birth. Without this buffering, Earth might have heated so much that all the water in its oceans boiled away and huge quantities of CO₂ accumulated in the atmosphere, much like Venus which has an average temperature of 462°C.  Or it might never have recovered from being a snowball, remaining permanently frozen.

Among the rocky worlds we know, Earth’s tectonics are unique. Venus and Mercury have no similar geological activity. Mars might have once, but not for billions of years. So why are we so lucky? 

According to geophysicist David Bercovici, of Yale University, models show the Earth sits right on the cusp between being a world with plate tectonics and one with a ‘stagnant lid’, like modern-day Mars or Venus. Something must have kicked it in the direction that produced a geologically active world that eventually gave birth to us. Bizarrely, even as astronomers probe planets hundreds of light-years distant, geologists still can’t precisely explain what triggered the events taking place beneath our feet. 

Tectonics derives from the Greek word ‘tektonikos’, meaning to build. It points to what we do understand about the way Earth’s surface is constantly remodelled. Our planet has a rigid shell called the lithosphere that comprises the crust and a hardened upper slice of an otherwise playdoh-like mantle (see diagram). That shell is cracked into seven large plates and a number of smaller ones that float on the mantle in slow, constant motion. 

The first inkling that continents moved dates back to the 1500s, when Flemish mapmaker Abraham Ortelius noted that the eastern and western coastlines of the Atlantic Ocean looked as if they might have once fitted together like pieces from a jigsaw puzzle. 

In 1912 German geophysicist Alfred Wegener coined the term ‘continental drift’ to describe how the lands on each side of the Atlantic had become so strangely sundered, but it wasn’t until 1963 that British marine geologists Fred Vine and Drummond Matthews provided the explanation (see Cosmos 54, p48). They realised the interior of the Earth is in motion. The rock of the mantle is slightly plastic – enough so that it can rise and fall in slow, roiling motions called convection currents: hot rock rises from the depths, cools, become denser and then descends. The best analogy is a lava lamp, which uses heat from a light bulb to induce the circulation of coloured wax in liquid. While the lava lamp’s convection currents are fast enough to produce mesmerising changes of colour, the rock of the mantle moves “about as fast as your fingernails grow”, says Bercovici – at a speed of less than 10 cm a year.

When rising currents hit the underside of the solid lithosphere, they deflect sideways, exerting drag. If that drag is strong enough, it can rip the lithosphere apart, creating new plates and making old ones move, upwelling magma filling in the gaps. When this happens at the bottom of the ocean, the result is ‘sea floor spreading’ – which is what Vine and Matthews observed. This is occurring today in places such as the Mid-Atlantic Ridge and the Red Sea Rift between Africa and Arabia. 

As the spreading crust cools, it grows denser. Eventually the leading edge furthest from the magma flow starts sinking back into the mantle, pulling the rest of the slab behind it – a process called subduction –  and so completing the convection cycle. Like the wax in the lava lamp, the cycle of rising, spreading, falling and rising again is the engine that moves the plates, and with them the continents, which ride atop like rafts. 

Though these motions occur at a rate of only a few centimetres per year, that is rapid enough to make even the oldest seafloor in the world startlingly young – less than 200 million years old. Continental crust, the buoyant crud that froths to the surface as ocean crust subducts, is much older.

The plates do not move in the same direction or at the same speed. This causes some plates to crash into each other, driving up mountain ranges, such as the Himalayas at the collision of the Indian and Eurasian plates. They can also grind past one another, as along California’s famed San Andreas Fault. Or one can dive beneath another, as occurs at the Pacific ‘Ring of Fire’ that circles the Pacific Ocean in a belt of earthquake-prone regions and volcanic activity. 

In this process, continents tend to remain on the surface. They are too buoyant to be easily subducted into the depths. But they still play an important role via a process known as ‘weathering’, which provides a vital thermostat that has helped keep the Earth temperate for billions of years.

It begins when CO₂ from the atmosphere dissolves in rainwater to form carbonic acid. This breaks down minerals in continental rocks, producing calcium and bicarbonate ions that wash into the sea. Marine organisms take them up to form calcium carbonate, the building block for their shells and skeletons, which ultimately settle to the seafloor and become limestone. 

Each year the process removes about 300 million tonnes of CO₂ from the atmosphere. But the carbon isn’t sequestered forever, because some of that limestone is subducted along with the seabed. It heats, melts and is incorporated into magma for carbon dioxide-spewing volcanoes to release. This also produces fresh rock for the next weathering cycle.

What makes this process function like a thermostat is that the more CO₂ there is in the atmosphere, the more carbonic acid there is in rain (and the more rapidly weathering occurs). This removes CO₂ from the atmosphere more swiftly, keeping the Earth from transforming into a Venusian runaway greenhouse. Conversely, if atmospheric CO₂ levels fall,weathering slows, allowing volcanic CO₂ to slowly build back up. It’s a slow, self-correcting process that for billions of years has kept the Earth’s temperature within a zone that is hospitable to life. 

So what got Earth’s plate tectonics going, rather than the planet ending up with a largely inert ‘stagnant lid’ like Mars and Venus?

The earliest Earth was all magma ocean with no solid surface to form plates, let alone plates that drift around and collide with one another. At a minimum, plate tectonics couldn’t have begun until after the Earth’s surface solidified, somewhere about 4.5 to 4 billion years ago. Just when the plate tectonics kicked in, though, still has geologists squabbling. 

If you’re seeking the earliest traces of plate tectonics, a good place to look is the Jack Hills in Western Australia. To the casual traveller this range of low mountains about 800 km north of Perth is not hugely impressive. But to geologists the hills are of towering significance, containing time capsules of the world’s oldest rocks in tiny crystals of zirconium silicate (ZrSiO₄). 

Zircons formed in cooling magma. Three things make them geological gems. First, they carry a date stamp of formation, based on the decay of traces of uranium trapped within them. Second, they are extremely durable; the ancient volcanic rocks that gave birth to them eroded long ago and were reconstituted into sedimentary rocks in the Jack Hills’ outcrops. Third, they bear trace elements like titanium and aluminium, which reveal the conditions of their birth. 

So far these zircon time capsules have telegraphed an extraordinary message: 4.2 billion years ago they were born kilometres below, crystallising as they rose to Earth’s surface. This tells us the mantle was starting to churn at that time.

But were these upwellings the same as those that drive modern plate tectonics? Craig O’Neill thinks not. He’s a cheery geodynamicist at Sydney’s Macquarie University who has been studying Jack Hills zircons for many years. In his view, the zircons could have been formed by localised upwellings similar to those occurring today in places like Hawaii and Yellowstone. In other words, not an Earth-wide tectonic churning but a local percolation.

Vicki Hansen, a planetary geologist at the University of Minnesota, Duluth, has come to the same conclusion based on “greenstone terranes” found in Greenland, South Africa, Canada and Scandinavia. 

These rock assemblages, which measure a few hundred kilometres across, date back to the Archaean Eon, 4 to 2.5 billion years ago. They are interesting because the greenish granites that give them their name are mixed up higgledy-piggledy with seabed sediments in ways we never see in more recent volcanic provinces. If modern-day rocks are like the vegetables displayed at the supermarket, the greenstone rocks are like stir-fry. This, Hansen says, indicates that whatever was going on in the Achaean involved processes “fundamentally different” to those today.

More evidence that modern plate tectonics had not geared into action until relatively recently comes from the study of the history of continental drift. 

If there’s any consensus among geologists, it is that something changed about 2.7 billion years ago to kick tectonic plates in action.

The most recent is Pangaea, which formed about 335 million years ago and lasted through much of the age of the dinosaurs. It was preceded by Rodinia (1 billion to 750 million years ago), then by Nuna (2 to 1.8 billion years ago). The earliest detectable supercontinent is Kenorland (2.7 to 2.4 billion years ago), relics of which are scattered across Western Australia, North America, Greenland, Scandinavia and the Kalahari Desert. 

The fact we can’t find a supercontinent older than Kenorland may simply mean the surviving bits are too scattered for geologists to piece back together. It’s like trying to figure out the history of a vase that has been broken and reassembled several times. 

But with supercontinent formation and break-up requiring modern-style plate tectonics, the fact we haven’t found one before Kenorland might instead be telling us that for the Earth’s first 1.8 billion years the lava lamp was not strong enough to produce anything other than localised percolations, not the continent-driving process we have today. 

If there’s any consensus amongst geologists, it is that something changed about 2.7 billion years ago to kick tectonic plates into action. “There appears to have been a major event,” says Kent Condie, a geochronologist at the New Mexico Institute of Mining and Technology in Socorro. 

But what could that have been? Theories range from the mundane to the dramatic, but all require the Earth to have overcome the same basic hurdles. Either the power of the lava lamp that makes mantle currents rise and swirl must have increased or the Earth’s crust must have weakened, allowing it to break into plates; or perhaps both occurred simultaneously.

One view, favoured by Matt Welller of Rice University, is that feedback loops in magma currents gradually built up to a level strong enough to produce self-sustaining plate tectonics via what engineers call a ‘hysteresis loop’. A hysteresis loop occurs when there is a lag between cause and effect. It is analogous to an out-of-tune automobile engine. When you press down on the accelerator, at first the engine barely reacts, then it lurches forward. 

Suppose the deep convection currents driving the Earth’s plate tectonics were to suddenly shut down. That would reduce the amount of heat that can escape, causing mantle rocks to heat up and become more plastic. Softer rocks can support more vigorous convection, so the lava-lamp effect intensifies, carrying heat more rapidly from the interior –until enough has escaped, the mantle cools and its currents slow again. 

“You can shift back and forth as you heat up and cool down, heat up and cool down,” says Julian Lowman, a geodynamicist from the University of Toronto. According to this view, the juvenile Earth experienced these on-and-off episodes on a small scale, producing the localised tectonics suggested by the Jack Hills zircons and the greenstone terrains. Then, about 2.7 billion years ago, these shifts became locked into a self-sustaining Earth-wide convection cycle.

Hansen, on the other hand, opts for a more dramatic scenario. The event that kicked off the tectonic plates might literally have been a kick – in the form of an asteroid or comet strike. Not as big as the one that formed the Moon, but far larger than the one that killed the dinosaurs. 

She first described her theory in 2007 in the journal Geology, arguing such an object would have punched right through the crust, heating the mantle and setting currents in motion, dragging the plates along with them and starting tectonic movements. Once plates began colliding and sinking, the process expanded until it spread across the planet. “Subduction is like a virus,” the paper states. “Once begun it can easily spread.”

Alternatively, the dramatic event might have come from below. In a 2015 paper in Nature, a team led by Teras Gerya, of the Swiss Federal Institute of Technology in Zurich, argued that hot spots on the Earth’s core could have caused plumes of hot mantle to rise beneath a continent. Under the right circumstances, they calculated, this could break up the continent and cause pieces to sink, creating a self-sustaining cycle that became plate tectonics.

Even the strongest mantle currents wouldn’t have triggered tectonic activity if the Earth’s crust was too strong to break into plates.

One might think a cooling Earth would have weaker tectonics. But it’s not that simple. “There are lines of research,” Lowman says, “suggesting that plate tectonics has a better chance of manifesting itself as a planet cools.” That’s because the lava-lamp engine that drives plate tectonics depends less on how hot the Earth’s interior is as on how rapidly it can transfer heat to the surface. The faster heat is transferred, the stronger the engine, and the stronger the mantle currents that drive tectonics.

It has been known since the 1930s that the Earth’s core has two layers: an outer one composed of molten metal, and an inner one made of solid metal. As the Earth cools, the inner core grows. In the process it releases heat energy – equal to the amount it took to melt all that material in the first place. That energy rises through the core, increasing the rate at which it heats the mantle and, ultimately, rises to the surface. 

 Supporting the theory that the cooling core may power the tectonic engine, a 2015 study by Condie and colleagues in the journal Precambrian Research traced the motions of continents over the past 2 billion years. They concluded that plate tectonics have been slowly speeding up, with average plate speed nearly doubling over that time. 

But even the strongest mantle currents would not have triggered tectonic activity if the Earth’s crust was too strong to break into plates. As it was, apparently, on Mars. For Berkovici, the key factor for the emergence of plate tectonics was therefore the weakening of Earth’s crust. It might have started gradually, beginning with the type of plume tectonics reflected in ancient greenstone terranes. Each of these magma breakthroughs would have created fault lines along which rocks slipped against each other, just as they do in today’s earthquakes. These motions would have produced weak spots that might have become focal points for later breakthroughs. Bercovici compares it to repeatedly bending a paper clip. “It gets softer,” he says. “Eventually you can bend it easily.” {%recommended 6333%}

Gradually these weak zones would have spread until they merged into plate boundaries similar to today’s, and the process went from local and intermittent to global and continuous. “A meteor might have gotten it started,” Bercovici says in a nod to Hansen, “but it needs these feedbacks to keep going.”

It is easy to try to fold all of this into a nice, coherent story. It would begin with a magma ocean, followed by weak, intermittent plume-style tectonics. These would eventually reach some tipping point that shifted the process to its present state, either due to changes in the core, an asteroid impact, the accumulation of Bercovici’s weak spots, or some combination of all three. But the plethora of options suggests caution. 

We may not yet have all the pieces to the puzzle. Lindy Elkins-Tanton, director of the School of Earth and Space Exploration at Arizona State University in Tempe, remembers being a graduate student at a conference, wondering what made scientists who disagreed with her own presentation so sure of themselves. 

“I sat there thinking perhaps I just didn’t know enough yet,” she recalls. “But now, 15 years later, I see that none of us know enough. We can only make small incremental progress in this very complicated problem.”