Snowball Earth: Antarctica As A Natural Laboratory

Photo Credit: Stephen Warren – Researchers explore the Mount Howe area in the Transantarctic Mountains during a previous field season. The blue ice area is similar to an ice type that likely existed during Snowball Earth events, a time of severe glaciation in the distant past.

By Peter Rejcek, Antarctic Sun Editor

Stephen Warren has made eight previous trips to Antarctica, including hunkering down for a winter at South Pole, to study its climate. But on his ninth visit to the frozen continent later this year, the professor from the University of Washington in Seattle will use Antarctic ice to learn more about Earth’s climate hundreds of millions of years ago.

“This is different,” he said. “Now we’re not trying to find out about Antarctica. We’re using Antarctica like a natural laboratory to find out about something else.”

That something else concerns the ice conditions that may have existed during a time dubbed Snowball Earth, when ice and snow blanketed the planet so completely that even the ocean surface at the equator froze.

It’s an intriguing and controversial hypothesis, said Warren, whose own research on the subject dates back some 30 years, even before the term Snowball Earth entered the lexicon. A term coined, incidentally, by Joe Kirschvink, a principal investigator on a current, unrelated National Science Foundation-funded project to study whether the mass extinction of the dinosaurs 65 million years ago began before an asteroid smacked into planet.

“It was a curiosity in climate models,” Warren explained of the Snowball Earth hypothesis that first intrigued him long ago. “It would appear in the climate models, and once you got in, it was so difficult to get out, we were sure it had never happened, because if it had happened we were sure the ice would still now be [here].”

In other words, once locked into a frozen ball of ice, no one could figure out how to hit the defrost button. It was Jim Kasting, now at Pennsylvania State University, who found the escape route, according to Warren.

Kasting pointed out in 1992 that carbon dioxide (CO2) emissions from volcanoes would build up a greenhouse effect after several million years sufficient to melt the ice. Kirschvink proposed in the same year that the snowball events actually had occurred, and Paul Hoffman of Harvard University later strengthened the hypothesis with geological and isotopic evidence.

There are other troubles with Snowball Earth. For example, it should have been traumatic for life — not only because of the extreme cold but the ice would have blocked photosynthesis — yet there’s no evidence of mass extinction. That has led some to argue that Snowball Earth is more like Slushball Earth, where severe glaciation occurred but not all the way to the equator. And certainly not with ice hundreds of meters thick on the tropical ocean.

Warren said climate models dictate otherwise when one factors the effects of low latitude glaciers that today sit thousands of meters above sea level. “If you cool the climate enough to get those glaciers to come all the way down to sea level, in a climate model, then the ocean freezes.”

It’s difficult, he added, to reconcile large areas of open ocean with sea-level glaciers. “The easiest conclusion is that the ocean was frozen,” he said.
Hot and cold running planet

Snowball Earth actually refers to a series of such extreme glaciations, the first dating back 2.2 billion years or so, about half the age of the Earth. Another series of snowball events occurred beginning about 700 million years ago, with the final event dated somewhere in the neighborhood of 570 million years ago, according to Warren.

The Earth was a much different place more than half a billion years in the past. A much-younger sun emitted about 6 percent less light than today. The supercontinent Rodinia had only recently broken up, leaving a jumble of landmasses concentrated at the equator. But what appears to have sent the planet into a deep freeze was a culprit most people are familiar with in a different context.

“It would have been a failure of the greenhouse effect,” Warren said. Just as today the pumping of carbon dioxide into the atmosphere is causing the Earth to warm up, the wholesale removal of the potent greenhouse gas drove temperatures downward.

What soaked up all that CO2? The process probably involved something called silicate weathering, according to Snowball Earth theorists. Atmospheric CO2 forms carbonic acid rain, which over geologic time weathers and disintegrates rocks and creates soil; rivers carry bicarbonate ions into the ocean as part of the normal carbon cycle, with CO2 eventually returning to the atmosphere by volcanic emissions, Warren said.

However, the weathering process can pick up speed during hot and wet periods, whether the cause of the heat is CO2 or some other greenhouse gas. Dan Schrag of Harvard University has found evidence that massive amounts of methane, an even stronger greenhouse gas, entered into the atmosphere prior to at least one snowball event, probably when plate tectonic movements depressurized a reservoir of methane locked in the sea.

The planet heated up, accelerating the weathering process and the removal of CO2 from the atmosphere. Remember, most of the land sat near the equator, so it would have been susceptible to weathering.

Meanwhile, methane has a relatively short lifespan in the atmosphere compared to carbon dioxide. Extreme cold and glaciation followed after the methane reservoir was finally exhausted and dissipated, with much of the CO2 removed from the atmosphere. And when the ice edge reached about 30 degrees latitude where solar radiation hits the tropics fairly uniformly, Warren said, “[the ice] jumps to the equator catastrophically.”
All about albedo

The runaway effect is thanks to the high albedo, or reflectivity, of snow and ice. The bright, white surface reflected the sunlight, and without much of a greenhouse to capture and store the heat in the atmosphere, the planet continued to cool and cool and cool. Temperatures probably reached about minus 30 degrees Celsius at the equator during Snowball Earth events, which may have lasted 10 to 30 million years.

The big thaw was a probably little less complicated. Without humans around to drive SUVs and pilot airplanes, the planet had to rely on a natural source to re-charge Earth’s greenhouse — Kasting’s volcanoes. The snow- and ice-covered planet at its extreme had very little in the way of a carbon sink, so carbon dioxide from volcanoes pumped CO2 levels back up.

“There’s no rain; no way to remove carbon dioxide,” Warren explained. “It builds up until you get so strong a greenhouse effect that you start melting the ice at the equator.”

Here is the more familiar scenario of today, where the bare ocean, with a much lower albedo, absorbs more sunlight and heat, creating a positive feedback of warming — much like what has happened in the Arctic in recent years. “It would have been a very rapid decline once it got started,” Warren noted.

[For more information about Snowball Earth and the possible scenarios for glaciation and de-glaciation, check out Hoffman’s Web site, originally funded by the National Science Foundation.]

One could be forgiven for thinking that ice and snow are pretty much the same when it comes to reflecting sunlight. However, Warren, whose expertise is the albedo effect, said one of the biggest uncertainties surrounding Snowball Earth concerns surface reflectivity.

“The reason for that is that the Earth would not have been totally snow covered,” he said. “There actually would have been quite a variety of surfaces.

“These ice types — if the snowball hypothesis is correct — used to exist on the tropical ocean. Now, they exist only in Antarctica. Some of them don’t even exist in Antarctica, so we grow them in the laboratory.”

Warren will lead a small team to look more closely at different ice conditions over the next two field seasons, starting with sea ice in McMurdo Sound during the Antarctic spring months of August and September. Conditions then should be cold and windy enough to create salt crystals in brine pockets within the sea ice.

“The ice gets brighter and reflects more sunlight,” said Warren, whose three-member team will collaborate with a New Zealand group that is now spending the winter in Antarctica working on the sea ice.

Investigations of other ice types will lead the team to the Garwood Valley later this season and then into the Transantarctic Mountains for the 2010-11 summer. His co-principal investigators, Bonnie Light and Edwin Waddington, will conduct laboratory experiments and modeling exercises to further examine the hypothesis.

The project, while illuminating the distant past, has implications for contemporary studies of climate here and on other planets, Warren said. “It’s intriguing for all kinds of reasons. It’s the most dramatic climate change to have happened in the history of the planet and yet it’s only been widely recognized in the last few years.”

The Snowball Earth hypothesis, some have argued, may have even spurred biological evolution, something Hoffman and Schrag pointed out in a Scientific American article earlier this decade. Though frozen over, small pools of open water may have existed at geothermal hotspots on coastlines, isolated for millions of years.

“That allows evolution to go on independently without competition from the other organisms developing in the other hotspots. It’s a way to promote evolution of diversity,” Warren said, adding that not long after the last snowball event, the Cambrian explosion, or radiation of complex animals, occurred.

“That’s when we first see macroscopic fossils of animals,” he said. “If there had been no snowball event, then life might still be happily microscopic on the Earth.”

NSF-funded research in this story: Stephen Warren, Edwin Waddington and Bonnie Light, University of Washington, Award No. 0739779.