Ancient meteorites and “fossilised raindrops” in lava show Earth’s atmosphere 2.7 billion years ago to be half the pressure of the atmosphere today with a distinct layer of oxygen in its top layer – a far cry from widely held assumptions that it was thick and completely oxygen-free.
In Nature today, Andy Tomkins from Monash University in Melbourne and colleagues from the UK and Australia report signatures of oxygen locked within sand-grain-sized meteorites that whizzed through Earth’s atmosphere 2.7 billion years ago.
And in Nature Geoscience on Monday, Sanjoy Som from the University of Washington and colleagues from the US and Australia examined bubbles trapped in ancient lava flows from the same period to calculate air pressure.
Both studies analysed samples collected from the Pilbara in Western Australia. The half-million square kilometre region is home to some of the world’s oldest rocks.
Tomkins didn’t start out using micrometeorites to measure the composition of the Earth’s atmosphere. He was originally interested in how the amount of space dust raining on Earth changed over time.
But after collecting micrometeorites embedded in limestone that formed around the end of the Archaean aeon (four billion to 2.5 billion years ago) near Port Hedland, he and his team saw the tiny space rocks contained certain types of iron oxide minerals – which, if Earth’s atmosphere was oxygen-free at the time, should not have existed. “I twigged to the fact that this was telling us about the chemistry of the upper atmosphere,” he says.
So how does oxygen affect a meteorite?
A fleck of rock, flying at 12 to 72 kilometres per second, melts as it encounters friction in the Earth’s atmosphere. (How much depends on the angle the micrometeorite approaches Earth, its size and initial speed.)
Melting lasts but a couple of seconds. It’s cool and hard again – or “quenched” – by the time it reaches the lower atmosphere. But in that short “hot phase”, molten minerals in the micrometeorite react with gases in the atmosphere and quenching locks in those reactions’ signatures.
Some micrometeorites landed in a lake and were kept pristine as limestone encased them. They were so well protected that even the Great Oxygenation Event, when photosynthetic cyanobacteria were thought to have pumped enormous amounts of oxygen into the atmosphere some 400 million years later, couldn’t reach them.
Analysis of the amount of iron oxidised in the micrometeorites showed the upper atmosphere’s oxygen levels were similar to those of today.
That oxygen probably came from carbon dioxide or sulfur dioxide that was split by sunlight (in a process called photolysis) in the upper atmosphere.
The results also suggest the upper and lower layers of the atmosphere 2.7 billion years ago did not mix. Previous studies of sediments from the same period show the lower atmosphere was all but devoid of oxygen.
And it also seems the atmosphere at sea level was far thinner than thought. Som’s study took advantage of bubbles trapped in lava that provide insights into air pressure.
As lava cools and hardens, dissolved gases bubble out into “fossilised raindrops”. Over millions of years, the bubbles gradually filled with minerals and look like white spots in the rock.
Bubbles at the top are bigger than those below because they don’t have to contend with the weight of the molten rock overhead – all that’s acting on them is the surrounding air pressure.
If scientists know the lava density and thickness, they can then compare the size of the bubbles at the top and bottom of a lava flow and calculate surrounding air pressure.
When Som and his colleagues did this, they found the air pressure 2.7 billion years ago to be half of modern levels – “the opposite of what we were expecting”, he says.
It was widely believed that the atmosphere must have been thick with nitrogen in order to hang on to heat pumped out by the younger, cooler Sun.
But with a thin atmosphere, greenhouse gases such as methane and carbon dioxide were probably more plentiful in order to retain heat.
The work also has implications for the climate of the time. For instance, water at such low pressures boils at around 58 °C.
“We’re still coming to grips with the magnitude of this,” says study co-author Roger Buick, also at the University of Washington.