About 717 million years ago, Earth’s wet landscapes and rolling blue waters turned into a cold, barren world. Scientists called this phase of geological history, and others like it, Snowball Earth.
What exactly froze the planet almost solid has been a mystery, as has how it stayed that way for 56 million years. On Wednesday, a team of researchers at the University of Sydney said they had figured it out. Earth’s glaciation, they say, may have resulted from a global decline in carbon dioxide emissions, as a result of fewer volcanoes spewing the gas into the atmosphere.
Less carbon dioxide makes it harder for Earth’s atmosphere to trap heat. If the depletion had been extreme enough, they argued, it could have pushed the planet into its longest ice age.
The theory, published in the journal Geology, adds to the understanding of how geological processes affected Earth’s past climate. It can also help scientists better understand trends in our current climate.
“These days, of course, humans have a big impact on CO2 in the atmosphere,” said Adriana Dutkiewicz, a sedimentologist at the University of Sydney who led the study. “But at the time, there were no people, and so everything was essentially modulated by geological processes.”
There are many ideas about what turned the Earth into a snowball. A popular theory suggests that minerals released by the weathering of igneous rocks absorb enough carbon dioxide from the atmosphere to cause a deep freeze.
Perhaps this helped start a global glaciation, said Dr. Dutkiewicz, but it couldn’t keep the Earth frozen for 56 million years on its own.
“So there must be some other mysterious mechanism that would have kept the glacier going for so long,” she said.
Dr. Dutkiewicz and her colleagues turned their attention to volcanoes because of a new model of Earth’s tectonic plates moving. As the continents drifted apart, they studied the changing length of the mid-ocean ridge—a chain of underwater volcanoes—predicted by the model.
The team then calculated the amount of volcanic gas emissions at the start and throughout the ice age. Their results showed a drop in atmospheric carbon dioxide sufficient to initiate and sustain a 56-million-year-old glaciation.
A reduction in volcanic gas emissions has been proposed as an explanation for Snowball Earth before. But according to Dr. Dutkiewicz, this is the first time researchers have proven that the mechanism was viable through model calculations.
Dietmar Müller, a geophysicist at the University of Sydney and an author of the study, said the work was a way “to distinguish between alternative models for this very ancient part of Earth’s evolution”. If scientists know there was an ice age, explained Dr. Müller, “then we can say that this reconstruction model is more likely than the other.”
Of course, a model is still just that: a model. Without real-world data to back it up, researchers can’t rule out other possibilities.
“One thing about geology, there are no set answers,” said Dr. Dutkiewicz. “But based on a combination of different lines of evidence, we can suggest that this is a very likely process.”
Francis Macdonald, a geologist at the University of California, Santa Barbara, who was not involved in the work, said studies like this were important for learning why climates fail. But he is reluctant to easily accept the results from ancient seafloor models because there are few records that reveal what the Earth’s oceanic crust was like at the time.
“How do we actually test it?” Dr. Macdonald asked about the team’s model. “I think it’s a really big challenge.”
However, Dr. Müller thinks it’s important to try to put limits on the amount of volcanic gas emitted in the past, especially when it comes to using climate models for the future. “Typically, this is the most uncertain parameter,” he said.
Research like this can help scientists distinguish the impact of geological activity from human-caused climate change. But could a natural decline in volcanic emissions ever save us from the amount of carbon we’ve poured into our atmosphere today?
“Unfortunately not,” said Dr. Dutkiewicz. “We can study these ancient concerns,” she added, “but human-induced change is a different kind of beast.”