Earth Science 1941

Canon of Insolation and the Ice-Age Problem

Milutin Milanković

Slow wobbles in Earth's orbit dim the high-latitude summer sun — and time the ice ages.

Choose your version

In depth · the introduction

The ice ages did not come and go at random — they kept time with the slow swaying of the Earth in its orbit.

The big idea

The Earth does not circle the Sun in a fixed way. Its orbit slowly stretches and rounds again; its axis tips a little more and a little less upright; and that tilted axis slowly swivels, like a spinning top, changing which season comes when the Earth is closest to the Sun. Each of these wobbles is gentle, and each repeats on its own long rhythm — tens of thousands of years.

Milutin Milanković showed that together they change how much summer sunlight reaches the far north. And it is the summer sun there that matters: when northern summers run cool, the previous winter's snow does not fully melt. Year after year it piles up, and an ice age begins. Warm summers melt it back, and the ice retreats.

How it came about

The idea had a Scottish forerunner, the self-taught janitor-scientist James Croll, who in the 1860s tied ice ages to the orbit — but guessed wrongly that cold winters were the cause. Milanković, a Serbian engineer turned professor in Belgrade, set out around 1911 to turn the whole thing into exact mathematics, for Earth, Mars and Venus alike.

It was a lifetime of arithmetic done by hand. He even worked on it as a prisoner during the First World War, continuing his calculations in a library in Budapest. The crucial steer came from the climatologist Wladimir Köppen, who — with his son-in-law Alfred Wegener, of continental-drift fame — told Milanković to focus on high-latitude summer. Their 1924 book carried Milanković's curve, and it matched the four known Alpine ice ages.

Why it mattered

For the first time the ice ages had a clock. Not a vague story of a colder past, but a precise prediction of when ice should have advanced and retreated, written in the geometry of the Solar System. It tied the history of our planet's climate to the motions of the heavens, and it gave geologists a calendar they could test against the rocks — and, eventually, against the mud at the bottom of the sea.

An analogy

Think of three dimmer switches wired to the summer sun over the far north. One is the tilt of the Earth, one is the shape of the orbit, one is the timing of the seasons. Each slides up and down on its own slow schedule. Most of the time their settings partly cancel. But every so often all three dim together — the northern summer goes faint and cool, the snow stops melting, and the ice creeps south. Milanković's achievement was to read the exact position of all three dials, for hundreds of thousands of years.

Where it sits

Milanković corrected and completed an idea older than himself (Adhémar, Croll) and handed it to the future. Rejected through the 1950s, it was confirmed in 1976 when deep-sea cores were found to pulse at exactly his orbital periods. Today it underpins how we date the last few million years of Earth history. It belongs with the Library's other keys to climate — Arrhenius on the greenhouse effect (arrhenius-1896) and Keeling's rising CO₂ (keeling-1960) — and with Wegener's drifting continents (wegener-1912), whose own collaborator Köppen helped point Milanković the right way.

The original document

Original source text

Milutin Milanković (1879–1958) · Kanon der Erdbestrahlung und seine Anwendung auf das Eiszeitenproblem · Royal Serbian Academy, Belgrade, 1941 (in German)

The cosmic problem

Milanković opens by setting himself a single, audacious problem: to compute, from celestial mechanics alone, exactly how much solar radiation each latitude of a planet receives in each season — and how that has changed over geological time. He solves it for the Earth and sketches it for Mars and Venus. The whole edifice rests on Newtonian gravitation and spherical astronomy; nothing is fitted to the geological record in advance.

The three orbital elements

Three slow changes in the Earth's motion govern the answer. The eccentricity of the orbit — how far from a circle the ellipse is — breathes in and out over roughly 100,000 and 413,000 years. The obliquity, the tilt of the spin axis, nods between about 22.1° and 24.5° over roughly 41,000 years. And the precession of the equinoxes — the slow swivel of the tilted axis, combined with the turning of the orbit's own long axis — shifts the season at which the Earth is nearest the Sun, with periods near 23,000 and 19,000 years.

The insolation integral

From these elements Milanković integrates the radiation arriving over a day and over a 'caloric' half-year. With Köppen's advice he fixes attention on the summer half-year at high northern latitudes, expressing each result as the latitude that would receive the same summer radiation today — the 65°N equivalent latitude. The governing quantity is the high-latitude summer insolation, not the annual total.

The curves of the past 600,000 years

He computes the summer-insolation curve back through six hundred millennia, entirely by hand. The criterion, again Köppen's: ice ages are born of cool summers that fail to melt the winter snow, not of cold winters. The dips in the curve are read as glacial epochs — and they line up with the four Alpine glaciations (Günz, Mindel, Riss, Würm) that Penck and Brückner had mapped from the field.

What lies beyond the astronomy

Milanković is candid that the orbital forcing is only the trigger. The amplification — growing ice that reflects more sunlight, shifting oceans and carbon — belongs to a physics he does not claim to have solved. The Canon supplies the metronome; the orchestra of feedbacks is left to others.

[ … ]

Belgrade, 1941