Snowball Earth

The original Snowball Earth hypothesis, as described in Wikipedia suggests that Earth was entirely covered by ice during parts of the Cryogenian period, from 790 to 630 million years ago. The theory was developed to explain sedimentary glacial deposits at tropical latitudes. Although the presence of glaciers is not disputed, the Snowball Earth hypothesis remains controversial. I'll let you read the articles above and below to make up your own mind.

I found at the Snowball Earth website an interesting figure plotting the ice line latitude vs Effective solar flux and CO₂ concentration. The figure shows three stable branches (sea-covered, partially ice-covered, and ice-free) linked by two unstable ranges.

The disconcerting factor from the figure is that in order to break free of the completely ice-covered branch, either the effective solar flux would have to increase by 25% or the CO₂ concentration would have to increase by almost three orders of magnitude over present values.

Fortunately we can investigate the effect of such a catastrophe with our model without re-arranging continents or destroying most of earth's lifeforms. The only change necessary to our reference atmosphere was a change in the albedo of the earth's surface, to 0.75, typical of clean snow and ice.

The resulting temperature profile (shown in blue on the plot below) indicates temperatures would be 70^o or more colder than the reference atmosphere. The average surface temperature would be -25.61^oC.

Clearly albedo does make a big difference. If the earth reflects 75% of the solar energy hitting it, it won't warm enough to radiate much long-wave IR into the atmosphere. We haven't considered the effects of water vapor, clouds, and ozone which might otherwise moderate the situation, but not enough as air at -40^oC can't hold much water.

snowball.html 5/22/08 Victor Frank