Catastrophic Events in the History of Life: Toward a New Understanding of Mass Extinctions in the Fossil Record - Part II

Author: Weinreb David B.
Institution: Molecular Biophysics and Biochemistry & Geology and Geophysics
Date: April 2002

In Part I of this series, we traced the historical development of a theory stating that the Cretaceous-Tertiary mass extinction resulted from the collision of a meteorite near the Yucatán Peninsula in Mexico. The idea of a K-T impact, first put forth by Luis and Walter Alvarez more than two decades ago, came under heavy criticism from paleontologists who had always thought the creatures who perished at the end of the Mesozoic were wiped out slowly over millions of years, not in a instant's time by a fiery meteorite.

By the late 1980s and early 1990s, the evidence supporting the Impact Theory was almost overwhelming. Many geologists and paleontologists began to wonder, "If it could happen once, could it happen twice?"

Here, in the second part of our series, we explore some very recent research that suggests a meteorite may have been the culprit not only in the K-T mass extinction, but also in some of the most devastating biological crises in the history of the earth.

The Death Star Hypothesis

How often do mass extinctions occur? While the Phanerozoic - an interval that lasted from 253 to 11 million years ago - was punctuated by five major extinction events, there have also been numerous minor events. The boundaries between geological periods (for instance, the transition between the Jurassic and Cretaceous) are synchronized with episodes of high extinction rates.

In 1982, David Raup and John Sepkoski, both of the University of Chicago, examined marine invertebrate biodiversity in the Phanerozoic. They divided this 242-million-year sequence into 39 equal-length intervals, and defined an "extinction event" as an interval where at least 2% of all known marine invertebrate families became extinct. Raup and Sepkoski observed that such extinction events occur about every 26 million years. Raup's initial reaction to the data was, "That can't be right." Surprised by the outcome of their analysis, the two researchers reasoned that there must be some explanation for this pronounced periodicity:

A first question is whether we are seeing the effects of a purely biological phenomenon or whether periodic extinction results from recurrent events or cycles in the physical environment. If the forcing agent is in the physical environment, does this reflect an earthbound process or something in space? One possibility is the passage of our solar system through the spiral arms of the Milky Way Galaxy, which has been estimated to occur on the order of 108 years. Shoemaker has argued that passage through the galactic arms should increase the comet flux and this could, following the Alvarez hypothesis, provide an explanation for the biological extinctions.

Raup and Sepkoski had boldly suggested the apparent periodicity in the marine extinction record was the result of cyclical astronomical events that vastly increased the probability of the earth being pelted by meteorites.

In 1983, Marc Davis and Richard Muller, of the University of California at Berkeley, proposed that the Sun has a yet-undetected companion star with an eccentric orbit. The "unseen companion" is generally about 2 light years away from the Sun, according to Davis. As the companion star passes through the Oort cloud of comets surrounding the Sun, it launches many of these comets in the general direction of the earth. Davis includes a few reassuring words in his 1984 Nature paper, noting that such a companion star, if it does indeed exist, will not launch another comet toward Earth for at least another 15 million years. Davis confesses that:

The major difficulty with our model is the apparent absence of an obvious companion to the Sun, and the existence of such an object is its most important prediction. We take this prediction seriously largely because of our inability to find any simpler explanation for the periodicity consistent with known facts.

However, there is indeed a simpler explanation: The periodicity observed in the marine fossil records is a statistical artifact. That is, extinctions don't really occur every 26 millions; it just appeared that way because of the way Raup and Sepkoski analyzed the data.

Silencing the Deathstar Hypothesis

Antoni Hoffman of Columbia University dealt a tragic blow to the Raup and Sepkoski hypothesis in 1985. Hoffman asked his colleagues to consider a sequence of two consecutive geological stages. There are four changes in species diversity that could occur in these stages: 1) There is an increase in species diversity in the first stage, followed by another increase in the second stage; 2) There is a decrease in species diversity in the first stage and a decrease in the next stage; 3) There is a decrease in the first stage and an increase in the following stage; and 4) There is a increase in the first stage and a decrease in the second stage. Raup and Seposki defined an extinction event by the occurrence of an increase in the first stage and a decrease in the second stage. Statistically, there is a one in four probability of getting such an extinction peak between any two stages. Moreover, the average length of each stage defined in the Raup and Sepkoski study is 6.4 million years. For every 6.4-million-year stage there is a 25% chance that the stage will end with extinction. Therefore, extinctions will occur, on average, every 25.6 million years. It is then merely a statistical coincidence that Raup and Sepkoski find extinction events spaced 26 million years apart.

The work of Raup and Sepkoski was an attempt to find an extraterrestrial cause for every extinction event in the marine fossil record. Their claim was that, whenever many species die-out suddenly, the culprit is a shower of meteorites that invading the earth. Although their findings were widely challenged, many researchers have remained optimistic that meteorite impacts may still hold the answer to not only the K-T extinction, but also the Triassic-Jurassic event and the most horrific episode of all, the Permo-Triassic mass extinction.

Manicouagan

About 212 million years ago, a gigantic meteorite may have struck in Quebec, Canada, creating the Manicouagan Crater. Unlike the Chixculub structure, Manicouagan is mostly exposed at the earth's surface. The structure is about 75 kilometers in diameter. The collision required to create such a crater must have released approximately 1022 Joules of energy - 100 times greater than the energy released if the world's entire nuclear arsenal were to be detonated simultaneously.

The Manicouagan impact was so forceful that it ejected material out of the atmosphere and sent it on a ballistic trajectory around the earth. Like the Chicxulub impact, the Manicouagan impact left behind a global geochemical signature in the rock record.

A little more than 200 million years ago, the earth's biota suffered a fairly significant mass extinction. Although more than 75% of species passed into extinction at the Triassic-Jurassic boundary, the event is not considered to be as devastating as the Cretaceous-Tertiary or the Permo-Triassic extinctions.

Did the Manicouagan impact result in the Triassic-Jurassic extinction?

Many paleontologists have argued that the two events - the impact and the extinction - were unrelated. The impact event occurred more than 10 million years before the end of the Triassic. According to the current view, an impact occurred, but had no pronounceable effects on the biosphere. Then, 10 million years later, the earth experienced a mass extinction, comparable in magnitude to the K-T event that would come 140 million years later. The situation here is basically the reverse of the K-T debate: early skeptics of the Alvarez theory cited the absence of a 65-million year old crater to claim the extinction was not brought on by an extraterrestrial cause. At Manicouagan, there is a crater, but no simultaneous mass extinction that it could have triggered.

Andrew Winslow, of the State University at Stony Brook in New York, claims the best explanation may simply be that the age assigned to the crater is incorrect, perhaps by as much as 10 million years. If the impact at Manicouagan occurred 200 million years ago, then there would be little debate as to whether it was responsible for (or at least one of the major causes of) the Triassic-Jurassic extinction. Since the precise date of the extinction event can only really be estimated to an accuracy of 5 million years or so, Winslow can make a good case for an extraterrestrial cause of the extinction by dating the crater to anywhere between about 198 and 204 million years ago.

Why are craters so difficult to date? The answer has to do with how they are formed. When the meteorite responsible for the Manicouagan crater struck Quebec nearly a quarter of a billion years ago, the force of the impact exhumed rocks that resided as deep as 9 kilometers below the surface. The surface of the newborn crater consisted of material brought to the impact site by the meteorite, rocks at the surface at the time of impact, and minerals exhumed from far below the earth's surface. At Manicouagan, some of the exhumed rocks may have been 800 million years old.

The heat generated by the impact liquefied the rock and, in effect, the crater became a melting pot for relatively young rocks at the surface and much older minerals originally buried kilometers below the site of impact. The heat released by the impact was so intense that it took between 1,600 and 5,000 years before the melted rocks cooled.

Geochemists generally determine the age of a rock by measuring the ratio of radioactive parent isotopes to radiogenic daughter isotopes. Lead is produced by three pathways: the decay of 238Ur (a process that occurs with a half-life of 4.5 billion years) to generate 206Pb; the decay of 235Ur to 207Pb (half-life of 704 million years); and the decay of 232Th to produce 208Pb (half-life of 14 billion years).

Isotope petrologists often rely on a mineral known as zircon to date rocks. Although zircons occur in very low abundances in most rocks, they are extremely resistant to weathering. When a zircon cools and crystallizes, it will not incorporate any lead into its crystal structure. Therefore, when geochemists shatter a zircon crystal, they know any lead found trapped within the crystal lattice has been produced by radioactive decay of uranium and thorium atoms previously trapped within the crystal. Theoretically, knowing the ratios of the various isotopes of uranium, thorium, and lead trapped within a zircon allows geochemists to calculate its age.

However, the techniques for dating zircons are not this straightforward. A zircon may be subjected to intense heat sometime after its initial formation, causing it to re-crystallize and exclude lead from its crystal structure, even lead atoms that may have been present in the structure prior to melting.

This is a big part of the problem at Manicouagan. Very old zircons (nearly a billion years old) buried deep within the crust were exhumed to the surface, where they melted together with much younger surface rocks. The intense heat from the impact caused the zircons to melt and "donate" their radiogenic lead isotopes to other minerals. These other minerals therefore have more radiogenic lead than would be expected for minerals of their age. For this reasons, the dates assigned to Manicouagan minerals are much too old - perhaps by 10 million years.

The same type of problem occurs in the Potassium-Argon dating system. Potassium-40 decays to the noble gas, 40Ar, with a half-life of 1.25 billion years. The argon remains trapped in the rock's crystal structure until some disruption - violent shaking as in the case of an earthquake, or intense heating - releases it. This process is known as degassing, and it can make rocks seem younger than they really are. For instance, a 70 million-year-old rock that has experienced degassing twice - for instance, at 45 and 25 million years ago - will be dated at only 25 million years old. This is simply because we are only measuring the argon formed since the time of the most recent degassing event.

The Manicouagan impact certainly would have triggered degassing of the radiogenic argon found within the very old rocks exhumed by the impact. The radiogenic argon released from these rocks may have been trapped within the structures of other minerals during the rapid cooling process that ensued. Normally, geochemists use the accumulation of radiogenic argon as a chronometer for how much time has elapsed since a rock cooled. However, under the extraordinary conditions of the Manicouagan impact, many rocks may have formed with radiogenic argon already trapped within their crystals. Winslow comments, "If this impact actually occurred 210-212 million years ago, and it didn't cause the Triassic-Jurassic extinction, then that would be incredible. How could something this big not cause a mass extinction?"

If we hope to attribute extinction events to meteorite or comet impacts, we need to learn to reliably date craters. The dating of such complex geological structures, which form under such extraordinary conditions, is far from simple.

The Permian-Triassic Extinction

If life on Earth has ever come close to being snuffed out entirely, it happened 245 million years ago. The Permian extinction was the most devastating crisis in the history of life. Although many explanations have been put forward to explain the rapid die-off of nearly 96% of all species, the fundamental underlying cause of the Permian-Triassic event is still largely debated.

In 1994, Douglas Erwin, of the Smithsonian Institution's National Museum of Natural History, suggested this mass extinction involved "a tangled web rather than a single mechanism." Global sea levels decreased at the end of the Permian, reducing the available habitats for many marine organisms. Erwin addresses the possibility that the eruption of the Siberian traps may have contributed to the end-of-Permian extinction. However, most of the basalts from the Siberian eruptions formed after the boundary (thus making their environmental effects an unlikely suspect in the search for a mechanism of extinction). Erwin warns fellow earth scientists:

Few complex events stem from a single cause; more common is a complex web of causality, a web that can be difficult to untangle. My own view is that the cause of the end-Permian extinction lies in such a tangled web. The most plausible explanation would appear to be a three-phase model combining elements of several mechanisms described previously. The extinction began with the loss of habitat area as the regression dried out many marine basins. The increased exposure of Pangea as the regression progressed exacerbated climatic instability. This instability, coupled with the effects of continuing volcanic eruptions and an increase in atmospheric carbon dioxide (with some degree of global warming), led to increasing environmental degradation and ecological collapse .The final phase of the extinction occurred in the earliest Triassic. The rapid transgression destroyed near-shore terrestrial habitats.

Certainly, it seems possible to explain the Permo-Triassic event without invoking an extraterrestrial cause as proposed for the K-T and T-J boundaries. However in February 2001, Luann Becker of the University of Washington published compelling evidence that an impact had indeed occurred in the final hours of the Permian.

Becker extracted a type of organic molecule known as fullerenes (C60 to C200) from clay sediments from Permo-Triassic sites in China, Hungray, and Japan. Fullerenes are molecules composed of 60 or more carbon atoms arranged to form a spherical, hollow, cage-like structure. These fullerenes, nicknamed "buckyballs" for their round shape, are constructed from 60 carbon atoms and are common in clays from the Permo-Triassic sites as well as Cretaceous-Tertiary sites; they are also found in meteorites. Extraordinarily large fullerenes, constructed of as many as 400 carbon atoms, were found in the 4.6-billion-year-old Allende meteorite that crashed in Mexico in the 1970s.

Inert gases, including helium and argon, are often trapped inside the cage of the buckyballs. Becker has hypothesized about the origin of the buckyballs themselves by measuring the ratio of helium isotopes trapped inside these gigantic carbon molecules. The helium isotopic composition of the fullerenes from the Permo-Triassic localities is similar to the signature in carbonaceous chrondrites, suggesting that the buckyballs were delivered to Earth in a meteorite or comet. The ratio of helium isotopes is dramatically different from that found in the earth's atmosphere, for instance. Becker and her colleagues propose that the buckyballs found at the Permo-Triassic sites were transported to Earth by the meteorite that triggered the most catastrophic extinction of all time.

Conclusion

While many questions remain unanswered, we now have come to recognize that the history of life has been profoundly affected by meteorite impacts. The 1980 Alvarez theory, that a meteorite impact brought about the demise of the dinosaurs, stunned most paleontologists and ultimately compelled them to re-examine the nature of the driving forces of evolution. Today, researchers have announced discoveries strongly suggesting that the mass extinctions at the Permo-Triassic and Triassic-Jurassic boundaries, among others, may also have been caused by meteorite impacts. The riddle of evolution demands that we search for answers perhaps beyond the realm of our imaginations. Life as it exists in our modern world is the product of 600 million years of metazoan evolution - and the earth's biota will continue to evolve, following a course yet to be chartered, for eons to come. As Stephen Jay Gould states in his book Wonderful Life, "The pageant of evolution is a staggeringly improbable series of events, sensible enough in retrospect and subject to rigorous explanation, but utterly unpredictable and quite unrepeatable."