When you look up at the moon's pockmarked face, you're actually staring at Earth's early history. The rain of asteroids that pummeled the lunar surface hit our planet too — it's just that erosion and plate tectonics blotted out the evidence. In fact, no rocks anywhere in the world survived to tell the story of the first 500 million years of Earth's 4.5-billion-year existence, a tumultuous period of frequent impacts known darkly as the Hadean.
Now, scientists have capitalized on the moon's long memory to uncover Earth's own past. The researchers found that much of our planet's surface probably melted repeatedly following large collisions during the Hadean eon. Some of these impacts likely vaporized the oceans and sanitized the planet of any early life that may have gained a foothold, according to a new study published Wednesday in the journal Nature.
While scientists have long recognized that large and frequent impacts shook the Hadean Earth, the new study marks the first attempt to quantify what might have happened.
The researchers started out by translating recent estimates of the cratering history for the moon — published over the last few years by the same group of scientists — into similar estimates for the Earth.
"The reason is very simple: If you have a crater, you had an impact," said Simone Marchi, a geologist at the Southwest Research Institute in Boulder, Colo., and lead author of the Nature study.
Marchi and his colleagues studied the frequency of different sized craters on the bare lunar surface to determine how often rubble of various shapes and sizes came hurtling through the solar system. They dated the impacts using moon rocks collected during the Apollo missions and by modeling how the number of incoming asteroids might have changed over time.
That worked well for objects up to 100 miles across, Marchi said.
"Above that size, we cannot use the moon for making a prediction simply because larger objects did not collide with it," he said.
So they turned to the Main Asteroid Belt, a cloud of debris between Mars and Jupiter that probably would have formed another planet if not for Jupiter's powerful gravitational pull.
The Asteroid Belt may not have been the source of early projectiles, but it provided a good match for the size distribution of smaller objects that hit the moon. Marchi and his colleagues figured it would also provide a sensible approximation for larger asteroids, too, so they used it to estimate the frequency of massive impactors.
The research team pulled these findings together to come up with a model for what would have happened as these objects collided with the young Earth — including how much melting would have occurred.
They found that almost every spot on Earth melted at sometime during the Hadean. Some places may have melted multiple times, and that melting may have extended down to a depth of 12 miles. This helps explain why no ancient crust survived, Marchi and his colleagues wrote.
But is their model correct?
"It's very difficult to test that because, of course, we don't have an impact record," said James Day, a geologist at the Scripps Institution of Oceanography in San Diego, who was not involved in the study.
Instead, the study authors compared their results to the few threads of evidence that hold clues about early Earth history, such as how much gold, osmium and similar elements reside in the mantle. If these elements — known as siderophiles, or iron-loving elements — were part of Earth's original composition, they would have migrated to the iron-nickel core. The fact that they exist in the mantle suggests they were probably injected into Earth's outer layers by asteroids later, providing a rough estimate of how much mass was added over time.
Marchi and his collaborators found that their model matched these observations if the Earth encountered dozens of projectiles larger than 100 miles wide during its first billion years of existence. What's more, these impactors could have ranged up to 2,500 miles in diameter — more than twice the size of the largest body in the Main Asteroid Belt today.
A few of these behemoths likely prowled the early solar system during the Hadean, left over from its formation, the authors said, and would easily have wiped out nascent life.
The researchers also compared their results to Hadean zircons, tiny indestructible crystals that comprise the only direct vestiges of early Earth. A handful of samples older than 4 billion years have been found in Canada and Australia, where they have been recycled through numerous violent iterations of the rock cycle. The scientists say their model explains the curious observation that, for some reason, most zircons from the early part of Earth's history date back to a narrow window of time between 4.1 billion and 4.2 billion years ago.
The authors hypothesize that older zircons may have formed during large asteroid impacts but then melted or had their internal clocks reset by the extreme pummeling and constant reworking of the early Hadean. Only the zircons created at the end of the bombardment period would have survived to tell the tale, and provide a date for the last volley of impacts.
However, not everyone buys the new story. T. Mark Harrison, a geologist and zircon expert at UCLA, says the argument does not fit the geologic evidence.
"They fundamentally misunderstand what a zircon is going to do," Harrison said. He thinks the chemistry of the crystals suggests they formed in contact with oceans at subduction zones such as the west coast of present-day South America, not under miles and miles of molten crust.
Although it may seem like a minor disagreement, Day, of Scripps, says it's a fairly important point: Any explanation for the early Earth must also explain the zircons.
"If you can't get them to fit your model, then your model probably doesn't work," he said.
Day said the new paper offers an interesting perspective on early Earth, but agrees that the authors' take on zircons is "probably pushing it a little far."
"It's inevitable that there were asteroid impacts at this time," he said. "Their size and scale are what they try to quantify, but our knowledge very sketchy."