MK ATLAS· Science Atlas

Milutin Milanković

Why do ice ages come back?

Earth Science· 1920· Belgrade · Milutin Milanković· Unsettled

1min readUpdated 2026-09-29

That time, that place

In the nineteenth century geologists found the marks of glaciers all over Europe: this place had been under ice.

And not once. The ice had come and gone repeatedly, and nobody knew why.

Why this question

Milanković spent part of the First World War as a prisoner. Under house arrest in Budapest he began thirty years of calculation with paper and pencil.

He saw that Earth's orbit is not perfectly steady. Three things drift very slowly: how elliptical the orbit is (about 100,000 years), how far the axis tilts (41,000), and how the axis itself gyrates like a top (23,000).

Where these coincide, the summer sunlight reaching the northern hemisphere rises and falls. A cool summer leaves last winter's snow unmelted, and a few thousand years of that makes an ice sheet.

What was found

Geologists were unimpressed. The change in sunlight seemed far too small to drive so large an effect.

In 1976 oxygen isotope ratios measured in deep-sea sediment cores gave a temperature record spanning hundreds of thousands of years. The cycles in it were 100,000, 41,000 and 23,000 years.

The three numbers he had worked out with a pencil were written in the mud at the bottom of the sea. Fifty-odd years after publication, eighteen years after his death.

The old idea

Nobody could say why ice ages recur

The evidence

The three predicted cycles appeared in deep-sea isotope records

What followed

Climate became calculable from orbital conditions, the backdrop to today's climate science