Skip Navigation

Did stardust trigger snowball Earth?

February 9, 2005 By Philip Ball This article courtesy of Nature News.

Clouds of interstellar molecules may have plunged our planet into a deep freeze.

Please log in to rate this page.

View Comments

Our planet may have frozen over in the past as it drifted though giant dust clouds in space. The result of the dust-bath would have been an almost complete overcoat of ice for the world, according to a new theory.

A group of US and Russian researchers argue that interstellar dust might have accumulated in Earth's atmosphere and cooled the planet, tipping the climate towards a 'snowball Earth' event in which ice sheets keep growing until they cover almost the entire globe.

But the idea does not persuade some geologists. "It conflicts with the geological record," says Daniel Schrag, a geochemist at Harvard University in Cambridge, Massachusetts. He points out that there seem to have been dramatic changes in the Earth's carbon cycle up to a million years before known snowball Earth events, which the dust-cloud hypothesis is at a loss to explain.

Alexander Pavlov, of the University of Colorado at Boulder, and his colleagues counter that their climate-cooling mechanism is almost inevitable, however. They say that on at least two occasions in the past 2 billion years, the Solar System must have passed through clouds of dust thick enough to cause a snowball Earth1,2. They think it is possible that two such ultracold episodes, 600 million and 750 million years ago, might have been triggered in this way.

Dusty answer

Snowball Earth events are much more severe than normal ice ages. They occur through a runaway process in which growing ice sheets reflect ever more sunlight back into space, resulting in further cooling and more ice. Eventually, the ice advances from the Poles virtually all the way to the Equator, trapping the planet in a deep freeze.

There is strong evidence in the geological record that Earth may have iced over in this way several times during its history. Various causes have been proposed, but Pavlov and his colleagues say that none is fully convincing.

They argue that their dust trigger is more plausible. Our Galaxy contains many giant molecular clouds, which are huge clusters of molecules that can clump into dust grains. As the Solar System moves through the galaxy, it passes through such clouds roughly once every 100 million to 1 billion years.

Pavlov and colleagues have calculated how much of this dust might be captured by Earth's gravitational field, filling the atmosphere with dust. Dust particles reflect sunlight, but they let Earth's heat out into space. In other words, they act as the precise opposite of greenhouse gases, cooling the planet.

On reflection

Such a cooling effect was observed after the eruption of Mount Pinatubo in the Philippines in 1991, which scattered volcanic dust into the atmosphere. The researchers calculate that the cooling effect of a passage through a dense molecular cloud could be at least two or three times greater.

That, they say, would be sufficient to trigger snowball cooling. If the planet were already on the verge of an ice age, even a molecular cloud of modest density could push it over the edge a larger freeze. The snowball Earth could then persist for about 10 million years, much longer than it would take the Solar System to cross a typical molecular cloud. The ice would thaw only when enough greenhouse gases from volcanoes had built up in the atmosphere.

The researchers suggest that there could be a detectable geological signature of such an event. Interstellar dust is enriched in the isotope uranium-235, relative to its natural abundance on Earth. This dust would gradually settle out of the atmosphere and find its way into sedimentary rocks laid down at the time of the snowball freeze.

Schrag doubts that such evidence, if it were to be found, would be conclusive. And he does not see how an extraterrestrial trigger for the cooling can explain the apparent timing of such events. "Why would you get two of them close together [600 and 750 million years ago], and then nothing?" he asks.


  1. Pavlov A. A., Toon O. B., Pavlov A. K., Bally J. & Pollard D. Geophys. Res. Lett 32, L03705 (2005).
  2. Pavlov A. A., Pavlov A. K., Mills M. J. , Ostryakov V. M., Vasilyev G. I. & Toon O. B. Geophys. Res. Lett 32, L01815 (2005).


User Tools [+] Expand

User Tools [-] Collapse

Pinterest button


Please log in to add this page to your favorites list.

Need Assistance?

If you need help or have a question please use the links below to help resolve your problem.