Milutin Milankovitch was a Serbian mathematician who, back in the 1930s, developed the idea that the Earth's long term climate was governed by slight variations in the motion of the Earth.
There were three variations that he considered. The first was the precession of the spinning Earth. Precession is the slow "wobble" of the Earth's axis, which is the sort of thing that can be seen in a spinning top. One wobble takes about 26,000 years. And the result is that, while the Earth's axis is pointing at the North Star, Polaris, right now, over the next few thousand years it will describe a circle in the star field, before coming back to Polaris in about 26,000 years time.
The second variation was the obliquity or tilt of the Earth's axis. This is the angle at which the Earth precesses or wobbles. This also changes by a few degrees in a 41,000 year cycle.
And finally there is the eccentricity of the Earth's orbit around the Sun. Planetary orbits are elliptical, and the eccentricity of an orbit is the degree to which it deviates from a perfect circle. The Earth's orbit is very nearly circular, but it becomes a bit more elliptical before returning to nearly circular again in a 100,000 year cycle.
Milankovitch calculated all these slight changes, and also the resulting changes in the amount of sunlight warming the Earth at different latitudes due to the changing precession and obliquity and eccentricity of the Earth. And he argued that when the warming by sunlight at 65 degrees N was at its greatest, glacial ice covering the northern hemisphere would be at its minimum, and that when the warming from sunlight was at a minimum, glacial ice would be at its maximum. Or that was the rough idea.
But the changes in the amount of sunlight are really quite small, because the variations in the Earth's precession and obliquity and eccentricity are rather small. It didn't seem to be enough to cause ice ages.
Neither did it seem to fit very well with ice core data (for example from the Antarctic Vostok ice core) which was subsequently used to estimate ice volume and temperature and CO2 concentrations over the past few hundred thousand years.
At right, the graph shows the variation in the sunlight warming at 65 N over the past 400,000 years as calculated by Milankovitch, ranging from around 460 W/m2 to 540 W/m2 per day - an anomaly of +/- 40 Watts. Below it are the temperatures (in degrees C) over the same period derived from the Vostok ice cores, with the peak temperatures corresponding to ice minima like in our current interglacial period.
And it's a bit hard to see much correlation of one with the other.
The result is that, while Milankovitch's theory is intellectually very attractive, its popularity has waxed and waned. Most often, while people concede that the cycle might have some small effect, it had to be amplified in some way to explain the cycles of glaciation over the past few million years. Global warmists have argued, for example, that the Milankovitch cycles might cause a little bit of warming, which resulted in CO2 being released from the oceans to cause more warming, ultimately flipping the Earth's climate from a deep glaciation to a warm interglacial.
But today I learned that a Washington University researcher, Gerard Roe, suggested a few years ago that people were looking at Milankovitch's theory all wrong. They were supposing that when the Milankovitch cycle was at its peak, this would correspond to an ice minimum, and when it was at its lowest, that would correspond to an ice maximum. What they should have been looking at was the change in the ice volume. For at the peak of the Milankovitch cycle, it wasn't that the climate was warm, but that it was getting warmer, and the ice was melting. As Roe put it:
So Roe went back to the ice core data, and worked out the rate at which the glaciers were expanding and contracting. And then he plotted the Milankovitch cycles over (on top of) the rate of change of volume of the glaciers.
It's an almost perfect match. And, according to Roe, there's no need to invoke any CO2-driven warming. In fact, he doubts that CO2 has much effect at all.
And it figures. The changes in the amounts of sunlight might be small, and wouldn't melt very much ice. But just a single slow drip of water of melt water would have added up, over tens of thousands of years, to an entire kilometre-thick ice sheet over half of Europe.
Roe's paper only came out a bit over 3 years ago, and physicist Lubos Motl, for one, only found out about it about 6 months ago. So it's a bit of news that's still slowly filtering out into the world. It's obviously not going to be reported by the global-warming-obsessed mass media. So the only way you'll ever hear about it is through the internet.
And it also illustrates beautifully how science is a process of finding things out, and how "the debate is never over", and how there will always be people like Gerard Roe who come along and point out something obvious, which everybody else had overlooked (including Milankovitch himself, who seems to have thought that the 40,000 year obliquity cycle would predominate).