Skip to main content

Iron formations in South African RocksNicolas J. Beukes, U. Johannesburg, South Africa

In the annals of life on Earth, few events rate as high in importance as the rise of an oxygen-rich atmosphere. A lethal toxin to early lifeforms, oxygen transformed the planet by enabling the basic chemical reaction that today powers most life as we know it – including us.

Now a new analysis has pushed back the clock on oxygen's ascendancy. While the conventional view is that the crucial gas started showing up in significant amounts 2.3 billion years ago, the latest results suggest it happened far sooner, some three billion years ago.

The revelation – which faces strong skepticism from some experts – is more than a matter of numbers. Oxygen is a byproduct of life itself, specifically of bacteria that convert sunlight into food via photosynthesis. The emergence of this complex biological process three billion years ago, not so long after life appeared, would mean microbial evolution operated more quickly than widely believed. It also means that as-yet unknown factors must explain why oxygen did not reach higher concentrations in Earth's atmosphere 2.3 billion years ago, when evidence for its presence is unequivocal.

"We were quite surprised," said Sean Crowe, a biogeochemist at the University of British Columbia and lead author of the study, published on Wednesday in the journal Nature. "We were expecting to find no evidence for oxygen at all."

Dr. Crowe said he and his collaborators made their discovery while testing a new technique they had developed for measuring oxygen exposure in rock and soil samples based on minute traces of chromium and other metals. For purposes of comparison, he asked a colleague in South Africa to send him samples from a geological formation so old that it was believed to have formed hundreds of millions of years before atmospheric oxygen made its debut.

Instead, their method indicated oxygen was present at roughly 0.03 per cent of its current concentration. Such a small amount would mean instant suffocation to a human being, but "a finding that there was any free oxygen in the environment at three billion years ago is significant," said Brian Kendall, an assistant professor of geochemistry at the University of Waterloo, who was not involved in the study.

If oxygen did make an early appearance, albeit a muted one, the result has implications for scientists seeking signs of life on planets elsewhere in the galaxy.

"One of the big questions is how often does [oxygen-producing] photosynthesis arise, and when it arises, does oxygen immediately go up?" said Jim Kasting, an astrobiologist at Pennsylvania State University.

Space telescopes are increasingly making it possible to analyze the atmospheres of planets orbiting distant stars by studying the way they absorb and reflect light. A strong indication of the presence of oxygen would be taken as a sign of life on another world, but the new analysis suggests oxygen-producing life could appear without astronomers being able to find it.

In an analysis published last May, Dr. Kasting reviewed several ways oxygen levels might have been kept extremely low on Earth – through volcanic activity, the movement of continents or geochemical reactions on the seafloor, for example – despite the early presence of bacteria capable of making it. But Dr. Kasting added he was not sure what to make of the results of the new study.

"It might mean that oxygen [levels] went up and down a few times," said Dr. Kasting. Others have made similar claims, but none have put the appearance of oxygen as far back as three billion years.

"It is a difficult challenge, because there are relatively few areas of the world where pristine rocks of the correct age are preserved," Dr. Kendall said.

The matter is unlikely to be settled soon. Joseph Kirschvink, a professor of geobiology at the California Institute of Technology, called the new results unconvincing and the technique unproven. He added that other studies of the same formation using other markers yielded no sign of oxygen at such an early date and he wondered whether the indications of oxygen exposure in the new study were more recent than they appeared.

"We live in an aerobic environment," Dr. Kirschvink said, "and you're looking for indicators of aerobic effects."

Dr. Crowe said he and his colleagues are gathering more samples for analysis in an effort to better pin down their results.