Yesterday, in A Theory of Ice Ages, I argued that when the Earth gets covered with ice, the ice acts as a layer of insulation exactly like the clothing on a human body. The underlying rock heats up, just like your skin warms up after you put on some clothes. And eventually the warm underlying rock may melt the ice above it. And when the layer of insulating ice is removed, the rock beneath starts to cool down again, just like your skin cools when you take your clothes off.
And so the cycle of ice ages is one of rock warming and cooling. During the ice age when the Earth is covered in ice, the rock beneath the ice heats up. And when the ice melts, the rock cools down again. And during the ice age there is a slow loss of heat from the rock, and during the interglacials between the ice ages, there is a high heat loss. And because the heat flows are so small, it takes a long time for the rock beneath the ice to warm up and melt the ice above it, And it takes a long time for the rock to cool down again. In fact it takes thousands of years.
We’re currently living in an interglacial period that started about 12 thousand years ago. And you might imagine that the ice just melted and hasn’t returned since.
But actually that wasn’t quite what happened. What actually happened was that the ice melted at the end of the last ice age, and then about a thousand years later it refroze. And it stayed frozen for another thousand years, and then melted again. And this brief return of the ice is called the Younger Dryas. (Dryas is an alpine flower, and I suppose that geologists must find dryas flowers in both the Younger Dryas and the Older Dryas.)
The Younger Dryas appears as a brief sharp dip in the temperature record below:
And it seems to be a bit of a mystery how the ice could melt and almost as quickly refreeze, before finally melting again. The chart above is taken from The Intriguing Problem of the Younger Dryas by Don Easterbrook, in which he writes:
The cause of these remarkably sudden climate changes has puzzled geologists and climatologists for decades and despite much effort to find the answer, can still only be considered enigmatic.
Is there any possible simple explanation of this in terms of heat flows like those I was considering yesterday? I think there might be.
In the first place, 15,000 years ago saw the end of an ice age that had lasted about 100,00o years. And that means that the rock beneath the ice had been slowly warming up for 100,000 years. And it had quite possibly got very hot. I’m not going to guess what temperature it reached, but I can well imagine that it could well have been hot enough not just to melt ice, but to boil water as well. There may well have been steam coming out of cracks and crevasses in the ice sheet.
These hot rocks would seem to have been melting the overlying ice pretty rapidly. And about 15,000 years ago, it seems they melted it all.
So what happened then? Well, the hot rock now became exposed to the cold air in the atmosphere, now that it had lost its layer of insulating ice. And the air temperature during the ice age seems to have been much colder than it is today. And so the hot rock would have started losing heat very rapidly to the cold air. And the effect of this was to warm up the air, and cool down the rock.
In fact the temperature of the rock could have plummeted very rapidly as it lost heat. So rapidly, that within a thousand years or so, it had fallen below the freezing point of water.
And when that happened, the ice started to rapidly build up again. The Younger Dryas had begun.
So now you’ve got cold rocks underlying the ice. And these rocks start to heat up again, now that they’ve got a blanketing layer of ice over them again.
But this time, instead of heating up slowly, the cold rock heats up rapidly. And it heats up rapidly because while the surface rock may have fallen below freezing, the rock beneath it was still very hot after those 100,000 years beneath the ice.
And so, in next to no time (about a thousand years), the ice that had just formed started melting again. And pretty soon it had all melted. And the Younger Dryas came to an end.
But, unlike at the end of the ice age, the rock that emerged into the upper air wasn’t quite as hot as it had been. It had lost a lot of heat during the first brief interglacial. And so when the the second interglacial period began, the rock was a lot cooler. And maybe the air in the atmosphere was warmer as well. And so this time, the rock lost heat rather more slowly to the atmosphere. And it didn’t cool down so rapidly as it had the first time.
And, crucially, it didn’t fall below freezing. So the ice didn’t start reforming again to create what would have been, I suppose, the Even Younger Dryas.
The rock temperatures at various depths below the surface are shown below (a bit roughly) with dark red for cold rock, light red for hot rock:
During period A the ice age is coming to an end, and there are very hot rocks under the ice. During period B, the ice has vanished, and the underlying rock is cooling as it loses heat to the atmosphere. Eventually the surface rocks fall below freezing point, and the ice reforms at the start of period C, the Younger Dryas. And then the underlying surface rock rapidly heats up again during the Younger Dryas, because there are still very hot rocks lying deeper beneath it. The ice then melts at the end of the Younger Dryas, but not quite as quickly as it did at the end of the ice age, because the rocks beneath the ice are cooler than they were back then. And we then enter the current modern (Holocene) period D, where the surface rocks are warmer than they were in the first interglacial B, but they are slowly cooling.
In summary: At the end of the ice age, the underlying rocks were very hot, and the atmosphere very cold, and this made for very large heat flows, and very rapid accompanying temperature changes – from freezing to warm to freezing and finally warm again. But by the end of it all, the hot rocks had cooled a lot, and the heat flows were smaller, and temperatures changed more slowly.
Anyway, that’s my guess. And it’s just a guess. I haven’t tried to build a heat flow model of it yet.