EARTH is 4.5 billion years old. It formed at the same time as the sun and the other planets in our solar system when thousands of rocks, large and small, collided and merged. Then came a hammer blow. A smaller planet seems to have struck Earth with such force that our world’s entire surface melted and huge amounts of material were blasted into orbit, eventually forming the moon. This explains why Earth’s first 500 million years are called the Hadean aeon after the mythical Greek underworld. But eventually, the vast sea of hot magma cooled and solidified into rock. Oceans formed. Land masses emerged.
The young planet would still have looked profoundly different from today. There was no oxygen in the air, so lots of metallic iron was dissolved in the water, staining the oceans green. Levels of atmospheric methane were perhaps high enough to colour the sky orange. And any land would have been barren rock, mostly dark black and grey.
Advertisement
How this otherworldly landscape came into being has long been a mystery. When did the oceans emerge? How deep were they? Was there always exposed land above the waves, or was Earth once a true water world? Now, a sketchy picture is starting to emerge. It doesn’t just reveal surprises about the making of our planet. Understanding when Earth first got dry land also shapes ideas about how life began. Some scenarios for its origin require exposed surfaces, so they can’t be true if everything was underwater. The question of Earth’s first land is therefore also a question about the origins of every living thing, including us.
There are almost as many hypotheses about the origin of life as there are scientists thinking about it. One much-discussed idea is that life began in alkaline vents on the seabed, where warm, chemical-rich water flows up into the ocean from deep inside Earth. This fits with evidence that the earliest microorganisms fed off chemical energy. But many chemists doubt that the building blocks of life, including amino acids and nucleic acids, could form in such places. A second school of thought holds that life emerged on land in a pool or pond. Experiments reveal that many key biomolecules can form in such a setting, often driven by cycles of wetting and drying. However, critics say the chemical reactions used in these studies throw up a problem, as they often require highly reactive substances that kill living organisms. Furthermore, supporters of the vent hypothesis sometimes argue that the pool hypothesis can’t be true on the young Earth. Yet that depends on when life emerged, which is also much contested.
The geological record gets increasingly spotty the further back in time you go. There are no rocks preserved from the Hadean and few from the next aeon, the Archean, which ran from 4 billion to 2.5 billion years ago. Nevertheless, there is confirmed evidence of microorganisms from 3.5 billion years ago preserved in rocks at Pilbara in Australia. In 2017, researchers led by at the University of New South Wales in Sydney, Australia, that these microbes lived in and around . Not everyone agrees with this interpretation, but there is undisputed evidence of microbes on land from 3.2 billion years ago. This comes from rocks in South Africa that record an estuary environment, complete with wind-blown sand dunes.
Still, this leaves a gap of at least a billion years from when Earth formed. In theory, life could have originated in a world-spanning ocean over 4 billion years ago: there is certainly no shortage of of earlier evidence of life. If so, say proponents of this view, whenever the first land emerged it was promptly colonised from the sea. But it is also possible that land formed much earlier than 3.5 billion years ago and that life originated there, not in a primordial ocean.
The sparse geological record makes these opposing ideas difficult to test. Until fairly recently, it was thought that the Hadean magma ocean lasted a long time, perhaps 500 million years. But two studies published in 2001 challenged this view. Both analysed zircons, tiny crystals that form in certain kinds of molten rock, whose chemical make-up holds clues about the environment where they formed. Researchers examined zircons from the Jack Hills in Australia, some of which were . One solitary crystal was even more ancient: . Yet all of them had a characteristic mix of different kinds of oxygen atoms, suggesting the rock from which they formed was interacting with liquid water. Later analyses indicated that Earth’s magma ocean had than 4.4 billion years ago. It would appear that the Hadean aeon wasn’t as hellish as was previously assumed.
The zircon findings have been taken to mean Earth had oceans 4.4 billion years ago, barely 100 million years after the planet formed. Van Kranendonk says that is unwarranted: “There does seem to be water-rock interaction, but you get water in magmas all the time, that’s not unusual. It doesn’t necessarily mean it’s the oceans.” He suspects the young Earth was still on the hot side, so instead of oceans of liquid water, there was a lot of steam and water vapour in the air. Still, oceans seem to have appeared earlier than we thought. Zircons show a around 4.2 billion years ago, which Van Kranendonk says marks the formation of large oceans.
Water world
Where all this water came from isn’t entirely clear. One idea is that the rocks that formed Earth contained water, which subsequently welled up to the surface. Alternatively, the newborn planet may have been dry and only received water later from incoming comets and asteroids. A third possibility is that the water came from deep space and was caught in Earth’s gravity. The debate continues, but what is clear is that water was plentiful from early on.
Whether this water took the form of a surface ocean or sank into Earth’s interior is another question – one that at Harvard University and his colleagues set out to answer in 2021. They compiled data on the amount of water that could be stored in rocks found in Earth’s mantle, the thick layer beneath the crust. It turned out that hot mantle rocks can’t carry as much water as when they are cooler. Earth’s interior was hotter just after it formed, so the young planet couldn’t store as much water in its interior as it does now. Today’s mantle has the potential to hold 2.3 times the mass of the ocean, but in the Hadean and early Archean, it could . That means some of the water that is currently inside Earth must have been on the surface when the planet was young. Dong estimates that the early ocean may have been as much as four times the volume it is today.
During the Hadean, “it’s possible the Earth was flooded with water, with almost no or little land”, says Dong. To at least some extent, this was carried through into the Archean. He points out that most of the rocks preserved from the early Archean don’t look like they came from dry land. “A lot of those rocks were formed under the water,” he says. Still, not much land doesn’t mean no land. After all, the Pilbara microbes seemingly lived on land 3.5 billion years ago. And there is even older evidence of land from the Isua Greenstone Belt in south-west Greenland, which contains rocks that are 3.7 billion years old. Most of them formed underwater, but some sections appear to be the .
The difficulty is that there are very few rocks from the early Archean, so if land was scarce, we shouldn’t necessarily expect to find any preserved. However, in recent years, geologists have found other lines of evidence. One was in a study published in February by researchers led by at the University of Bergen in Norway. They looked at rocks called barites found in southern Africa, India and Australia, that formed between 3.5 billion and 3.2 billion years ago in the middle of the Archean. Barites contain two versions of the element strontium. One type is more common in oceanic crust, the other in continental crust. By measuring how much of each is preserved in the rock, the team determined whether the seas were receiving sediments from the continents as a result of weathering by natural processes such as wind – and thus whether large land masses existed.
Roerdink’s team found that chemical weathering of land rocks was gradually increasing between 3.5 billion and 3.2 billion years ago. Extrapolating back to earlier times, they calculated that weathering of land . “We cannot say there was no land before 3.7 billion years ago, but what we can say is that from around that time, weathering of this land started to significantly affect the chemistry of the oceans,” she says.
This makes sense in the light of new evidence about what was happening in the young Earth’s crust at that time (see “The birth of plate tectonics“). As the planet cooled, its crust formed into the sort of large, mobile plates that we see today. If these tectonic plates started jostling around and bashing into one another some 3.8 billion years ago, that would explain why Roerdink’s team found evidence of significant land masses shortly afterwards, at 3.7 billion years ago.
The researchers also estimated how much land was needed to account for the weathering they observed. Whereas today about 29 per cent of Earth’s surface is land, between 3.5 billion and 3.2 billion years ago, it was just 2 to 12 per cent. It took hundreds of millions of years for these early continents to grow to something approaching modern sizes. The first supercontinent probably started forming .
Land ahoy
Huge uncertainties remain, but the evidence for when dry land existed has been pushed back a long way. “My feeling is that certainly since 3.7 billion years ago, and probably before that, there have been exposed land masses at the Earth’s surface,” says Van Kranendonk.
All this points to a tale of three parts in the making of Earth. Early in the Hadean, the molten planet acquired a steamy atmosphere with limited surface water. Then oceans formed and Earth was almost or entirely submerged. Finally, starting in the early Archean, plate tectonics got going so that isolated islands were gradually joined by larger land masses.
What does this mean for the origin of life? It seems that large oceans formed around 4.2 billion years ago, so if life arose in a deep-sea vent it must have been after that. But if life began on land, it could have appeared earlier, when water first started falling onto the surface, as much as 4.4 billion years ago. “As the atmosphere is raining out, then you actually have the whole surface of the planet as a geochemical reactor,” says Van Kranendonk. Alternatively, life could have emerged later, on volcanic islands, which probably existed 3.7 billion years ago and perhaps earlier.
Although we still can’t pinpoint the time or place of life’s origins, we do now have a much better picture of the early life of our home planet. From steamy magma blob, to water world, to the Archean Earth of green seas, orange skies and black rocks, it may all seem quite alien. But Van Kranendonk doesn’t see it that way. Within a billion years, there was already a lot that we would find familiar, from the vast oceans to the land masses dotted with volcanoes and hot springs, he says. “From my perspective, I feel like I can recognise early Earth.”
The birth of plate tectonics
Today, Earth’s crust is divided into several dozen large tectonic plates, which move around at speeds that are imperceptible to the naked eye. Over millions of years, the plates change positions and continents are carried from place to place. When plates push against one another, one may be forced down or “subducted” into the mantle.
The newly formed Earth was too hot for the crust to form into rigid plates. “The mantle might have been too hot and the crust too buoyant and too squishy,” says at Harvard University. What happened instead is much debated. One possibility is that Earth : the crust became divided into plates, but they . Alternatively, material may have accumulated on the underside of the crust and dripped off into the mantle when it got too heavy – a process called sagduction.
Estimates of when “modern” plate tectonics got started – from to . “That’s a huge range,” says Drabon. In a study published in April, she and her team tried to pinpoint when the plates started moving. They studied the chemical make-up of South African zircons from between 3.3 billion and 4.1 billion years ago. Chemicals in these tiny crystals hold clues about the rocks they formed in, and the team found evidence of a around 3.8 billion years ago. Older zircons were formed in very long-lived crust, suggesting little or no subduction was happening. But younger zircons carried evidence of short-lived crust, indicating that tectonic plates were moving and being recycled. This may or may not have been true subduction, but Drabon calls it “the first step toward the system that we have today”.
Other researchers have found unambiguous evidence of subduction, but rather later, around 3.2 billion or 3 billion years ago. It may be that there were isolated instances of subduction and other modern-like activity early on, perhaps partly triggered by asteroid impacts, with global mobile tectonics only taking hold later.
