Snowball Earth: The Cryogenian Global Glaciations Explained

Snowball Earth is the name for two extraordinary ice ages, the Sturtian and the Marinoan glaciations, that gripped the planet between about 717 and 635 million years ago, during a span of time geologists call the Cryogenian Period. Rocks from every continent suggest that glaciers reached sea level near the equator and that ice may have covered almost all of the oceans. The episode is one of the most extreme climate events in Earth’s history, and its aftermath set the scene for the first large, complex organisms.

Background: a world before animals

In the late Neoproterozoic, roughly 750 million years ago, Earth looked very different from today. There were no land plants or animals. Life consisted of bacteria, single-celled eukaryotes and simple algae, most of them living in the sea.

The continents were also arranged differently. Most land had been gathered into the supercontinent Rodinia, which was breaking apart from about 800 million years ago. Many of the fragments lay at low latitudes, near the equator.

The Sun was also around 6 per cent dimmer than it is now, because stars like the Sun slowly brighten over their lifetimes. That alone did not freeze the planet, but it meant that Earth had less margin against a large drop in greenhouse gases.

How the idea developed

Geologists had noticed puzzling glacial deposits in Precambrian rocks since the late nineteenth century. In the 1960s the British geologist W. Brian Harland argued that glacial deposits of about the same age appeared on almost every continent, including in places that magnetic evidence suggested lay near the equator. He proposed a great “infra-Cambrian” glaciation.

At about the same time, the Soviet climatologist Mikhail Budyko built simple climate models showing that ice can reinforce itself. Ice and snow reflect sunlight, so if ice spread beyond a certain latitude, the planet would cool further and the ice would race towards the equator. Budyko thought such a runaway freeze would be permanent and therefore had never happened.

In 1992 the American geobiologist Joseph Kirschvink coined the term “Snowball Earth”. He suggested that a frozen planet could escape because volcanoes would keep pumping out carbon dioxide while the normal processes that remove it were shut down. He also pointed out that iron-rich rocks reappearing at this time could be explained by an ice-covered ocean starved of oxygen.

In 1998 Paul F. Hoffman, Daniel Schrag and colleagues published an influential paper in Science based on fieldwork in Namibia. They argued that unusual carbonate rocks and carbon-isotope patterns above the glacial deposits fitted Kirschvink’s model, and the hypothesis became one of the most debated ideas in Earth science.

The two great glaciations

The Sturtian glaciation

The first and longer episode is named after the Sturt River near Adelaide, South Australia, where its glacial deposits were described. Precise dating of volcanic rocks in north-western Canada places its onset at about 717 million years ago.

It lasted until roughly 660 million years ago, making it more than 50 million years long. Its start coincides closely with the eruption of the Franklin large igneous province, a vast outpouring of basalt across what is now Arctic Canada. Many researchers think the rapid weathering of these fresh basalts in the tropics drew down enough carbon dioxide to tip the climate into a freeze.

The Marinoan glaciation

After a warmer interval of perhaps 10 to 15 million years, ice returned. The Marinoan glaciation is named after Marino, a suburb of Adelaide. It was shorter, probably lasting at least several million years, and ended about 635 million years ago, a date fixed by volcanic ash layers in Namibia and South China.

In South Australia’s Flinders Ranges, the glacial Elatina Formation is capped by a thin pink carbonate layer. In 2004 the base of that layer at Enorama Creek was chosen as the starting point of the Ediacaran Period, the first new geological period to be defined in more than a century.

The evidence for a frozen world

Glacial deposits near the equator

The key evidence is glacial sediment. Diamictites are jumbled rocks containing pebbles and boulders of many sizes set in fine mud, typical of debris dropped by glaciers. Dropstones, isolated rocks that fell from floating ice into soft seabed mud, are also common.

Measurements of ancient magnetism locked into the rocks, called palaeomagnetism, show that some of these deposits formed within a few degrees of the equator. Glaciers at sea level in the tropics imply a world far colder than any modern ice age.

Cap carbonates and carbon isotopes

Almost everywhere, the glacial deposits are sharply overlain by “cap carbonates”, layers of limestone or dolostone that normally form in warm water. Their sudden appearance on top of ice-age debris suggests an abrupt switch from deep freeze to hothouse conditions.

The ratio of carbon isotopes in these carbonates also swings sharply before and after the glaciations. Hoffman and Schrag interpreted these swings as a sign that biological productivity in the oceans collapsed under the ice and that carbon dioxide from volcanoes accumulated in the atmosphere.

Iron formations

Banded iron formations, layered rocks rich in iron, had largely disappeared from the record more than a billion years earlier. They reappear in Cryogenian glacial deposits, for example in the Rapitan Group of north-western Canada. Iron dissolves in seawater only when there is little oxygen, so their return fits an ocean sealed off from the air by ice.

How Earth escaped

On a frozen planet, rainfall and rock weathering slow dramatically, and so does photosynthesis in the sea. These processes normally remove carbon dioxide from the air, but volcanoes kept erupting. Over millions of years carbon dioxide built up, perhaps to hundreds of times today’s level according to some estimates.

Eventually the greenhouse effect overcame the reflective ice. Once melting began, darker open water absorbed more sunlight and the process accelerated. Models suggest the final deglaciation could have taken only a few thousand years in geological terms, producing a hot, humid world with intense weathering.

That weathering washed huge amounts of dissolved calcium and bicarbonate into the oceans, which helps to explain the thick cap carbonates. Some cap carbonates contain unusual structures, such as tubes and giant ripples, which researchers link to this violent change in climate.

Debates: hard snowball or slushball?

Not all scientists accept that the oceans froze completely. The “hard snowball” view holds that sea ice, possibly hundreds of metres to over a kilometre thick in places, covered virtually the whole ocean. Critics point out that photosynthetic life survived and that some deposits show signs of open water and active ocean currents.

Alternative “slushball” or “waterbelt” models propose that a narrow band of open or thinly frozen water remained near the equator. Others question whether all glacial deposits of similar age really formed at the same time. Better dating has, however, strengthened the case that the Sturtian and Marinoan glaciations were long, synchronous and global in reach.

The causes remain debated as well. Suggested triggers include the breakup of Rodinia, which created long coastlines and wetter climates; the weathering of tropical basalt provinces; and changes in the carbon cycle linked to living organisms.

Life and the aftermath

Life clearly survived. Microbial mats, algae and other eukaryotes probably persisted in refuges such as meltwater ponds on the ice, cracks in thin ice, hydrothermal springs and areas of open water. Chemical traces of algae called biomarkers suggest that algae became far more abundant in the oceans between the two glaciations.

Within tens of millions of years of the Marinoan thaw, the oceans held the first large, complex organisms of the Ediacaran biota. Microfossils from China’s Doushantuo Formation, which lies directly on Marinoan cap carbonate, show increasingly complex cells in this interval. Many researchers suspect a link between the glaciations, rising oxygen and the rise of animals, though how direct that link was is uncertain.

A later, smaller glaciation, the Gaskiers glaciation of about 580 million years ago, was not global. The animal radiations that followed culminated in the Cambrian Explosion.

By the numbers

Timeline

• c. 800-750 million years ago: The supercontinent Rodinia breaks apart, with many fragments in the tropics.
• c. 717 million years ago: The Sturtian glaciation begins, close in time to the Franklin basalt eruptions in Arctic Canada.
• c. 660 million years ago: The Sturtian ice melts, followed by a warmer interglacial interval.
• c. 650-640 million years ago: The Marinoan glaciation begins; its exact start is less precisely dated.
• c. 635 million years ago: The Marinoan ice melts and cap carbonates form worldwide, marking the start of the Ediacaran Period.
• 1964: W. Brian Harland argues for a great global glaciation late in the Precambrian.
• 1992: Joseph Kirschvink coins the term Snowball Earth.
• 1998: Paul Hoffman, Daniel Schrag and colleagues publish their Namibian evidence in Science.
• 2004: The base of the Ediacaran Period is fixed at Enorama Creek in South Australia’s Flinders Ranges.

Why it matters

Snowball Earth shows how far Earth’s climate can swing, from tropical glaciers to a hothouse, through feedbacks involving ice, rock weathering and carbon dioxide. It also sits at a turning point for life: the world that emerged from the ice was the one in which complex animals first appeared.