A Snowball's Chance June 30, 2000

Darwin was a disciple of uniformism: that the processes that occurred in the past are the same processes we see today. Evolution operates by the slow accumulation of changes driven by natural selection. But much of the work over the last 30 years or so has served to demonstrate just how important one-time catastrophic events can be in the history of life on Earth.

The discovery of iridium, a mostly extraterrestrial element, by Alvarez in the rocks at the Cretaceous-Tertiary boundary (about 65 million years ago), opened the door to our understanding of the most popular catastrophic event. An asteroid, which normally contains iridium, hit the Earth, the dust blotted out the sun and resulted in the extinction of all the large animals, mainly dinosaurs. This event has entered the popular culture so easily that it is seen as premises for blockbusters (i.e. Armageddon) and as rides in Disney World (an extremely fun one formerly called 'Countdown to extinction' but renamed 'Dinosaur' for marketing reasons). Yet, this event was not even one of the most devastating in the Earth's history.

The record for that is the Permian-Triassic event (about 250 million years ago). Over 90% of ALL life died out during that event. Many theories have been postulated for its cause, including a really massive asteroid or comet. Whatever, it was, it happened in a very short period of time. The rocks indicate that it took less than 150,000 years. But an approach using ¹³C indicates that it might have been less than 10,000 years.

The carbon isotopic technique examines the prevalence of ¹³C compared to ¹²C in carbonate rocks. Photosynthesis prefers to use ¹²C in its reactions. So plants will remove ¹²C from the CO₂ dissolved in the surrounding water. Thus, when carbon precipitates out of the ocean as carbonates, it tends to be rich in ¹³C if the oceans are teeming with life. No life and the carbonates will appear to be enriched in ¹²C. By examining the relative ratios of the 2 carbon isotopes, you can get an idea of how the carbon cycle is functioning and how life is doing. High ¹³C levels in the rock indicate that photosynthesis is working well by removing ¹²C from the water. But during periods where the carbon cycle is disrupted, more ¹²C finds its way into the carbonate rocks. So, using a known concentration as the baseline, scientists examine the change in ¹³C levels or the d ¹³C.

Well, during the Permian event, there was a very large, very rapid decrease in the d ¹³C levels in the carbonate rocks, indicating a large increase in the amount of 'light' carbon in the waters. Photosynthesis was not removing the ¹²C from the water. Something killed the carbon cycle and it did it very quickly. The spike is less than 10,000 years long, the limit of resolution in the dating process. This increases the probability that it was a traumatic event and not one just due to increased volcanism. The event killed off most life on Earth (the cockroaches were almost done in) but some life was still able to survive and eventually flourish.

The carbon cycle is a pretty powerful measure of the global climate. There are many other excursions of ¹³C in the record, and many of them occur at the sites of mass extinctions. However, several occured during times too early in the history of life to leave many fossil records. These occurred in the Neoproterozoic age, just before the Cambrian explosion of metazoan life, over 550 million years ago.

The climate of the early Earth was pretty wild. Ice ages would alternate with Steam ages. The variability in climate was greater than anything we have seen in the last 500 million years. In fact, there are good arguments that life itself has had a buffering effect on global climate. Interestingly, a recent theory proposes that life itself may have been responsible for some of the ice ages. Geologists have been studying many of the deposits from this time and noticed some really interesting things. One was that there were a lot of glacial deposits in tropical areas. The land mass of the Neoproterozoic age was concentrated in the Southern Hemisphere. Yet glaciers extended from the poles to the equator. How could this be?

The d ¹³C levels give a clue. During the late Neoproterozoic, there are several periods where high amounts of ¹³C are found in the rocks, increasing the d ¹³C levels. This would indicate that large amounts of ¹²C was being removed from the ocean, by photosynthesis, and that the resulting resulting carbonates would be rich in ¹³C. But what would happen if large amounts of CO₂ were removed from the environment?

If increasing amounts of CO₂ warm the Earth in the global warming we see today, guess what happens when CO₂ is removed? That's right. Global cooling. According to the original model, the surface of the Earth would freeze entirely due to a runaway process.Snowball Earth. Lower CO₂ levels would allow the polar caps to increase, which would reflect more light, increasing the size of the caps more. Simple modeling indicated that there would not be anything to stop this process until the Earth froze over completely, blocking out the sun to all parts of the ocean. Estimates from carbonate deposition indicate that the Earth remained in this condition for a minimum of 4 million years and a maximum of 30 million. This would certainly have a major effect of life on Earth. In fact, every positive excursion of d ¹³C in the Neoproterozoic is followed by a negative one, indicating that photosynthesis has broken down.

But this theory also helped explain the huge amounts of carbonates that were produced at the end of the age. The warming atmosphere and waters would cause dissolved carbonate to precipitate out. Both geology and modeling indicate that a pretty substantial event occurred that would potentially have horrific effects on life on Earth.What is intriguing is the possibility that early life may have triggered these changes. A Science News article helps clarify this (scroll down to find the very good figures explaining the theory). The slow break up of the landmass creates a large amount of shallow water regions. Ocean life explodes in the new environment, removing CO₂ from the atmosphere as it grows, leaving behind increased levels of ¹³C. Low levels of CO₂ result in a runwaway Snowball Earth. Most life dies but, without increased CO₂ levels, nothing can change. The Earth has reached a new equilibrium that would appear to be inimical to life. The Earth remains frozen until CO₂ levels rise, perhaps due to outgassing from volcanos. The increased amounts of CO₂ were then deposited to form the carbonate caps. Potentially, the existence of life itself lead to this world-wide calamity.

Of course, the idea of a frozen Earth has serious ramifications for life as we know it. So a more thourough examination using bigger computer models was just published. It is both more and less comforting. The simulations show that at CO₂ levels about 1/2 of today's levels, there is a very rapid transition, with a drop in average global temperatures from 10 °C to -25 °C. The time for this transition was about 2000 years, so fast on a geologic timescale as to be instantaneous. And the reverse transition, due to increasing CO₂ would be just as fast. How would life respond to such rapid changes, faster and more devestating than anything mankind could devise?

The models could also produce the same sorts of results with higher levels of CO₂ if the size of continental freeboard was increased. Then the 'snowball Earth' solution could be generated with CO₂ levels 1.5-2 times higher than today. Under these conditions, the glaciers on land would be over 5000 M thick in some spots. The temperature at Seattle's latitude would average only -30 °C or so with temperatures nearing -110 °C at the South Pole. But, at CO₂ concentrations about 2.5 times today's levels, they got a solution which still allowed open water to exist in the tropics. This could allow life to exist in conditions that are slightly better than a full snowball. So there might have been areas where eukaryoytic life could have survivied.

The authors suggest looking for such 'refugiums' for metazoan life. These areas would be Neoproterozoic land masses some distance from the main continent. From the map, areas in North China look to be the best shot. These areas could provide vital information regarding metazoan evolution. Because, even though no fossils of modern metazoans have been found from this time, data using molecular clocks indicate that several of the important genes used in metazoan development originated around this time. The creation of the modern body types that eventually led to us might have been dependent on several episodes of 'global cooling'. I'll talk some more next week about some of these molecular events. But what I find fascinating is the complementary interactions between biology, geology, paleontology and physics to clarify some of the most important, if ancient, events in the history of life on Earth. Maybe if the continental landmasses had looked different, there would not have been a frozen Earth to exert selective pressures that led to human beings. Perhaps we really do have a snowball's chance of existing on this planet.