Cobra Effect · Weather and climate

The climate model that passed the test of time

How a simple computer model predicted climate change decades before it arrived.

7 cards, read aloud in 2:40, with a test and sources.

A tall column of air divided into layers, beside an old computer cabinet

In the 1960s, Syukuro Manabe set out to build the climate inside a computer.

Manabe was a Japanese born meteorologist working in the United States. Computers were slow, so he started with the simplest model he could. He treated the atmosphere as a single column of air, from the ground to the top of the sky.

His column of air had to move heat in two ways at once.

Sunlight and invisible heat rays passed up and down through the layers. Warm air near the ground also rose, carrying heat upward, as it does on a summer afternoon. Getting both to work together was the key to a realistic column.

In 1967, with Richard Wetherald, he doubled the carbon dioxide in the model.

The air near the surface warmed by a little over two degrees. The air high above, in the stratosphere, cooled. That pattern is a fingerprint, because warming from a brighter Sun would heat both layers.

Later, Manabe and Wetherald built a model of the whole globe.

In 1975 it simulated the atmosphere around a simplified Earth, with winds, clouds and rain. It showed the far north warming faster than the rest of the planet. Much bigger computers would be needed to add real oceans, ice and land.

Decades later, the planet has behaved much as the early models said it would.

The lower atmosphere has warmed, and the stratosphere has cooled. The Arctic has warmed far faster than the global average. A 2019 study checked seventeen past model projections against what happened, and found most of them close.

In 2021, Manabe shared the Nobel Prize in Physics.

Half the prize went to Manabe and Klaus Hasselmann for the physical modelling of Earth’s climate. The other half went to Giorgio Parisi, for work on disorder in complex systems. Manabe was ninety years old.

So when a model makes a prediction, ask whether its past predictions came true.

A model is only a simplified copy of the world. Its value lies in whether what it predicts matches what is measured later. Manabe’s simple models passed that test decades after they were built. Starting simple, and checking against reality, is how a model earns trust.

Sources

Nearby ideas

All stories

Cobra EffectIdeas worth knowing
Streak 0 daysRead 0 of 0Known 0Kept 0
Every story is written to a rule: framing can be invented, facts cannot, and a tale that is probably a legend says so. Narration and pictures are generated once and kept. Your progress is saved on this device and in a record of its own on the server, so you can pick it up on another one. Nothing else about you is collected, and there is a button to delete the lot. Privacy.