The Orbital Clock Behind the Ice Ages

An idea worked out by hand

In the early twentieth century, while imprisoned and later working through the First World War, a Serbian engineer named Milutin Milankovitch set himself an enormous calculation. He wanted to compute, by hand, how the amount of sunlight reaching different latitudes of the Earth had varied over hundreds of thousands of years, driven by slow changes in the planet’s orbit and tilt. His hypothesis was that these changes in sunlight, small in the global average but large in their seasonal and geographic distribution, were what drove the great ice ages in and out. It took him decades, and for a long time the idea was treated as speculative. It is now one of the central pillars of how we understand the climate of the deep past.

Three slow wobbles

The Earth’s path around the Sun and its orientation in space are not fixed. They vary in three distinct ways, each on its own rhythm, and Milankovitch’s insight was that these combine to modulate the sunlight reaching the high northern latitudes, where the great ice sheets grow and melt.

Cycle	What changes	Period
Eccentricity	shape of orbit, near-circle to more elliptical	~100,000 and 413,000 years
Obliquity	tilt of the axis, between about 22.1° and 24.5°	~41,000 years
Precession	direction the tilted axis points, the slow wobble of a spinning top	~19,000 to 23,000 years

Eccentricity changes how round the orbit is, which affects how much the Earth-Sun distance varies over a year. Obliquity is the tilt of the axis, the thing responsible for the seasons; when the tilt is greater, seasons are more extreme, with hotter summers and colder winters. Precession is the slow wobble of the axis, like the gradual circling of a spinning top’s handle, which changes the time of year at which Earth is closest to the Sun and therefore which hemisphere gets its summer at closest approach.

Why summer in the north is the key

The crucial quantity, in Milankovitch’s analysis, is not the total sunlight averaged over the globe and the year, which barely changes. It is the amount of sunlight reaching the high northern latitudes during summer. The reasoning is about snow. Ice ages are not really about how cold winters are; plenty of snow falls in any cold winter. They are about whether the previous winter’s snow survives the following summer. If northern summers are cool enough that snow lingers year-round, it accumulates, compresses into ice, and builds the great ice sheets over millennia. If summers are warm enough to melt the winter snow, no ice sheet can grow. Since most of the land that can host large ice sheets is in the Northern Hemisphere, it is northern summer sunlight that holds the controlling lever, and all three orbital cycles feed into it.

From hypothesis to confirmation

For decades the theory lacked a decisive test, because no one had a climate record long and detailed enough to compare against the orbital calculations. That record came from the bottom of the ocean. The shells of tiny marine organisms, preserved in seafloor sediments, record the temperature and ice volume of their time in the ratio of oxygen isotopes they contain. By drilling long cores of sediment and reading that isotope ratio down through the layers, scientists reconstructed the climate of the past several hundred thousand years. In 1976 a landmark study by James Hays, John Imbrie, and Nicholas Shackleton analyzed such cores and found, written into the data, the very periods Milankovitch had calculated: cycles of roughly 100,000, 41,000, and 23,000 years, matching eccentricity, obliquity, and precession. They titled their paper to call the orbital variations the pacemaker of the ice ages, and the name stuck. The clock in the sky and the clock in the mud agreed.

The loose end

The theory is not entirely tidy, and honesty requires naming the main puzzle, known as the hundred-thousand-year problem. Over roughly the last million years, the ice ages have marched mainly to the slow 100,000-year beat of eccentricity. But eccentricity is the weakest of the three cycles in its direct effect on sunlight; on its own it should barely register. Why the climate responds so strongly to the faintest of the three drivers is still debated, and the answer seems to involve feedbacks within the climate system itself, the growth and collapse of ice sheets, changes in carbon dioxide, and the reflectivity of ice, which amplify a small orbital nudge into a large swing. The orbit sets the timing, but the Earth’s own machinery sets much of the size of the response.

That distinction, between a pacemaker and an amplifier, is probably the most important thing Milankovitch’s theory teaches. The orbital cycles do not directly freeze and thaw the planet. They tap out a rhythm in the distribution of sunlight, and the climate system, with all its feedbacks, dances to it, sometimes far more vigorously than the faint tap would suggest. A man working through orbital mechanics with pen and paper found the beat. It took ocean sediments, fifty years later, to confirm he had been listening to the right clock.