AN ENCOUNTER with a cloud of interstellar gas could plunge our planet into a deep freeze that could drastically impact life on Earth. An event like this may have been what triggered 鈥淪nowball Earth鈥� 鈥� a time when the entire planet was covered in snow and ice.
In 1992, Joe Kirschvink, a specialist in palaeomagnetism from California Institute of Technology in Pasadena coined the term 鈥淪nowball Earth鈥� to describe such a big chill. A few years later, Harvard University geologists Paul Hoffman and Dan Schrag strongly supported Kirschvink鈥檚 idea and argued that Earth was completely sheathed in ice at least twice in its history 鈥� about 600 million and 750 million years ago. But the cause has remained a mystery.
Now Alex Pavlov of the University of Colorado, Boulder, and his colleagues say that interstellar gas could be the culprit. The Milky Way is laced with gigantic clouds of hydrogen, which are concentrated mainly in the galaxy鈥檚 spiral arms. As our solar system orbits the centre of the galaxy it passes in and out of these arms, encountering about eight such clouds in the 250 million years that it takes to complete an orbit.
Advertisement
The damage would be done by the sprinkling of carbon and silicate dust particles these clouds contain. The Earth鈥檚 atmosphere is normally protected from this dust by the pressure of the solar wind 鈥� a blast of hot ions streaming out from the sun. But around 1 in 30 of these gas clouds are dense enough to overpower the solar wind, allowing dust to pour into the Earth鈥檚 atmosphere, Pavlov says (Geophysical Research Letters, vol 82, p 3705).
The influx would block as much sunlight as the outpourings from three volcanic eruptions a year, and it would last a lot longer. 鈥淚t takes 200,000 years to cross the cloud,鈥� Pavlov says. The result would be a Snowball Earth. Though Hoffman was initially cool towards Pavlov鈥檚 theory, he has quickly come round. 鈥淚 can鈥檛 see anything wrong with this idea,鈥� he says.
If dust did pour into the atmosphere, it might have left increased levels of uranium-235 in rocks. 鈥淚t鈥檚 for geologists now to go and look,鈥� Pavlov says. But Hoffman is not convinced that looking for uranium-235 is the right thing to do. 鈥淯ranium isotope ratios are easily altered in carbonate rocks,鈥� he says. He hopes to devise other tests to verify Pavlov鈥檚 theory.
In 2002 Schrag and Pavlov independently came up with another possible trigger for Snowball Earth: a fall in levels of the greenhouse gas methane. But Pavlov now says it is difficult to explain how methane levels could have risen again. 鈥淚 would love it to be true, but it doesn鈥檛 work,鈥� he says. Other explanations have also had trouble accounting for the abrupt change necessary to trigger a Snowball Earth.
鈥淭he influx of dust would block as much sunlight as three volcanic eruptions a year, and it would last a lot longer鈥�
While Earth will rarely encounter galactic clouds dense enough to trigger a global ice age, less dense clouds could also have a serious effect. By compressing the protective solar wind shell surrounding the solar system, they might allow galactic cosmic rays and hydrogen from the cloud to enter the atmosphere. This would destroy the Earth鈥檚 ozone layer and allow in ultraviolet-B radiation from the sun that would kill many species (Geophysical Research Letters, vol 82, p 1815).
Our last line of defence against this would be the Earth鈥檚 magnetic field, but this vanishes roughly every 200,000 years when the poles reverse. 鈥淚t鈥檚 not as bad as a snowball but it would probably destroy a lot,鈥� Pavlov says.
