Please, sno' more

I'm sure I'm not alone as I happily anticipate the first stirrings of spring here on the US east coast. I'm not a fan of winter. Or snow. Sure, it looks pretty for a while, but then it just gets in the way of every day life. I'm more than ready for this week's thaw.

I was contemplating all this as I sat on the bus yesterday morning, watching the piles of dirty DC snow on the sidewalks rush past me. I realised how thankful I am that we live on an Earth that is not covered with snow ALL the time. But then I remembered, our world hasn't always necessarily been like this.

The grey, gloomy, and snow-covered National Mall; slippery conditions prevail.

If we were to go back 650 million years, and then some, we would probably find an Earth with a very different climate. Geologists have long thought that this ancient period was one of a 'snowball Earth', when the surface was entirely (or almost) frozen.

The term for this hypothesis was first coined by Caltech scientist Joseph Kirschvink in 1992, but the idea of a global glaciation had previously been proposed to explain, for instance, glacial rock deposits in places such as Greenland. These rocks, known as tillite, are formed when glacial till—basically the unsorted crud that glaciers pick up, drag along, and then deposit as they melt—becomes lithified. Finding this kind of material in chilly places like Greenland shouldn't really be much of a surprise. But the rocks in question are old. Very old. And the process of continental drift means that Greenland's current position in Earth's northern latitudes was not its location when these rocks formed.

By studying the magnetism of certain minerals, which capture the direction of the surrounding magnetic field as they form, 'paleomagicians' are able to reconstruct the position of ancient continents around the globe. Such paleomagnetic studies on the tillites from Greenland show that they formed at tropical latitudes, where the Earth receives more (because of Earth's tilt on its axis) of the Sun's warming radiation—hardly where you might expect glaciers to exist.

Tillite deposit in East Greenland. Credit: M. Hambrey

And there are plenty of other distinctive rock types (e.g., banded iron formations and cap carbonate rocks) that date from this same era, which appear to be evidence of a global glaciation. But how could this state of Earth-wide freezing have occurred?

Climate models show that if sea ice advanced far enough towards the equator then a positive feedback system would have been established. The bright albedo (reflectance) of sea ice versus seawater means that more radiation is reflected away from the water in its solid form. So if the area of Earth's surface that was covered by ice increased, more light would have been reflected away and the Earth would have cooled. This would have led to more ice formation and more reflection and more cooling... You get the idea.

How Earth may have looked during a 'snowball Earth' episode.
Credit: geology.fullerton.edu

So if the snowball Earth really did exist, how did the planet get itself into such a state in the first place and how did it manage to escape?

To start the ice formation, obviously there had to have been some kind of large-scale initial cooling event. Options include: a supervolcano eruption that may have thickened the atmosphere and reduced the amount of radiation received from the Sun, or perturbations in the Earth's orbital dynamics (which follow the Milankovitch cycles) that brought the planet into a particularly cold configuration.

And greenhouse gases—namely carbon dioxide and methane—were probably the route for escape from Earth's frozen hell. In much the same way that the build-up of these molecules in today's atmosphere is to blame for global warming, the steady increase of their concentration in our ancient atmosphere could have allowed the huge accumulations of ice to thaw and melt. It has been estimated that about 350 times the concentration of carbon dioxide in today's atmosphere were needed to achieve such a feat. And it seems that volcanic eruptions, occurring over tens of millions of years, could have emitted these gases in the required quantities to melt ice in the tropics. This would have initiated an opposite feedback loop and would bring sea ice levels down to the more modest quantities of more modern times.

Although the snowball Earth hypothesis is still disputed, the fact that it immediately precedes the large-scale development of multi-cellular life—popularly known as the Cambrian explosion—is an intriguing fact. I think, therefore, at least one of the rock samples that serve as evidence for this hypothesis warrants being sent into space for another planetary geologist to recover. They are testament to the winter of our Earth's discontent and the conditions that had likely stifled the blossoming of life.

Bring on the spring. I eagerly await the sight of Washington DC in full blossom again.