Life goes on.
Given my current interest in ice ages, I naturally took a lot of interest in Youtube’s Snowball Earth, when some 700 million years ago there was ice covering the entire planet. According to Wikipedia:
Global temperature fell so low that the equator was as cold as modern-day Antarctica. This low temperature was maintained by the high albedo of the ice sheets, which reflected most incoming solar energy into space. A lack of heat-retaining clouds, caused by water vapor freezing out of the atmosphere, amplified this effect.
The whole Earth got covered in ice, and stayed encased in ice for some 50 million years. The most interesting thing to me was that, according to the Youtube video, the ice age ended with the eruption of “10,000 volcanoes”.
And this is what I would have expected. For once the entire Earth was covered in kilometres of ice, the surface rocks beneath the ice would have gradually warmed up. The centre of the Earth is currently believed to have a temperature of 7000ºC, and so the Earth’s mean temperature is maybe 2000ºC, and if no heat could escape through the ice, the Earth would have very gradually acquired that uniform temperature throughout.
And 2000ºC is well above the melting point of granite (about 1200ºC). So the surface rocks beneath the ice would have slowly climbed towards melting point. More or less the entire surface of the Earth beneath the ice would have turned into molten or near-molten rock. So of course there would have been 10,000 volcanoes. And they would have melted the ice pretty rapudly.
The same programme said that in the end-Permian ice age also ended with large scale volcanic eruptions. Volcanic eruptions at the end of ice ages should be expected.
But I think that when the surface of the Earth heats up by several hundred degrees, the surface will expand: most materials expand when heated. And this thermal expansion would have meant that rocks all over the surface of the earth would have been in compression as they pushed against each other, and the compression would have have been stronger the higher the surface rock temperature was. And when rocks become highly compressed at the surface of the Earth, they are likely to buckle or fracture, and the only direction they can buckle is upwards. They’ll form hills and mountains.
I wondered how compressed the rocks would become. The coefficient of linear thermal expansion of sandstone, α, is about 12 x 10-6, such that the length L of a rock is given by
L = Lo.( 1 + α.ΔT )
where Lo is the initial length, and ΔT the increase in temperature. And so if Lo was 1,000 km, and ΔT was 500ºC, the sandstone rocks along the 1,000 km length would have increased in length by 6 km. And if the crust was 1 km thick, a total of 6 cubic km of rock would have to buckle or get squeezed upwards to form a mountain. And given a 30º angle of repose of a pile of broken rock, the resulting mountain would be 1.86 km high.
And this suggests a variant of plate tectonics, one in which the forces acting to build mountains are the product of thermal expansion of surface rocks rather than the up-welling convection in the Earth’s mantle beneath the plate.
And these mountain ranges would lie roughly along N-S and E-W axes, because the Earth’s poles are cooler than its equatorial regiions, and so there would be a distinct N-S and E-W symmetry about the system of mountain ranges (the Andes being one N-S range of mountains and the Himalayas an E-W range of mountains).
And these expanding and contracting plates would push and pull each other around. During ice ages, the plates would warm up and expand and push against each other, and during interglacial periods when the surface rocks cooled down, they would contract and pull against each other. And this would produce flat plains like the US midwest or the Russian steppes. And if they pulled strongly enough apart, the rigid crust would fracture, and molten rock would well upwards to fill the gap (e.g. the mid-ocean Atlantic ridge).
The forces acting horizontally in these plates would be purely a function of their temperature and their lateral width Lo. Large plates would generate larger forces along their perimeters than small plates. So if the large Pacific plate has volcanic activity all around its perimeter (the “ring of fire”), it’s largely a consequence of its size.
Periods of orogeny (mountain-building) would be largely restricted to the ends of long and deep ice ages when surface rock temperatures had risen to sufficient temperatures for surface rock buckling to commence, and volcanoes to erupt through the broken rock. They would probably have been quite short and sudden. Entire mountain ranges might have been thrown up in days. And then when the compressive stresses had been relieved, very little mountain-building would take place until the end of the next major glaciation.
And life did go on through the Snowball Earth episode. In fact, the end of it brought the appearance of the first multicellular lifeforms. But that’s another story.