Back to Lecture 23


Lecture 24 - The Impact Theory of Mass Extinction

In previous lectures we looked at the range of dinosaurs that existed at the close of the Cretaceous. We can now turn our attention to the events around the great mass extinction event that involved the elimination of the non-avian dinosaurs as well as the associated extinctions.


Recall that in the 18th century Georges Cuvier worked on the stratal successions around Paris and believed that the history of life was marked by periods of creation, stasis, then obliteration. This catastrophic view of the history of life was countered by Charles Lyell's extreme uniformitarian view that very small changes added up over vast amounts of time were responsible for the seemingly great changes seen in the geological record. Lyell actually believed, at least early on, that there were no extinctions, just changes in abundances.

However, by the mid 19th century the basic shape of diversity of life was fairly well understood, and there did seem to be an overall pattern. In 1860, paleontologist John Phillips formalized what was known about diversity from the Cambrian Period on by recognizing (left) the Paleozoic Era (or ancient life), the Mesozoic Era (or middle life), and the Cenozoic Era (or recent life), with each era boundary marked by a period of lowered diversity - in other words by a period of accelerated extinction. The most abrupt of these reductions in diversity was the boundary between the Mesozoic and Cenozoic - that is between the Cretaceous and Tertiary Periods.

 Darwin, who applied Lyellian Uniformitarianism to biology, thought the fossil record was a bit of an embarrassment to his theory of evolution, especially in the apparent lack of intermediates. Darwin felt that the lack of intermediates was due to the imperfection of the geological record, and that they must be there - and of course intermediates between major groups, like Archaeopteryx, were soon discovered.

 The dual successes of Lyellian Uniformitarianism and Darwinian evolution lead to a view that persisted for the next century - that interpretations of the geological record in terms of discontinuities or catastrophes was to be more-or-less abhorred - and indeed there was precious little real evidence that this approach was not justified.

 However, by the 1960's and 1970's paleontologists began to carefully compile literature records of taxa from strata of different ages. It became apparent that there really were times of very high extinction rate.

graph of extinctions from SepkoskiA more recent tabulation (right) of shelly marine invertebrates by Jack Sepkoski of the University of Chicago shows that there have been several times in the last 270 million years (the better-preserved part of the record) when there were very high levels of extinction. In this graph, "percent extinction" is the number of extinctions of genera in an interval of time divided by the number genera present (or at risk) in that interval, times 100 (to make it a percent). Note the large peaks in extinction at the end of the Permian, Triassic, and Cretaceous periods.

 Particularly the Cretaceous-Tertiary (K-T) boundary stands out because there was considerable stratigraphic evidence that marine and continental extinctions looked like they might be synchronous. In specific the marine biota was hit very hard:

15% of all marine families (Sepkoski, 1982)
but 50 % at generic level
and maybe 80-90 % of all species.
The continental biota was also hit hard:
about 25 % at the family-level became extinct
but about 56 % at the genus
a Cretaceous foraminiferLet look now at some specifics, sort of a body count:

 MARINE ORGANISMS (particularly important ones in bold)

planktonic foraminifera 83%
ostracodes -50 %
sponges -69 %
corals -65 %
sea urchins -54%
ammonites -100%
marine reptiles 93 %
reptiles in general -56%
but non-avian dinosaurs and pterosaurs -100%
But many things also made it through or were relatively unaffected:
higher plants -10 %
dinoflagellates -5 %
mammals +120 %
These extinctions were clearly of great magnitude, but in the intellectual milieu of the time, there was much speculation, but little specific work on the nature of the K-T boundary.

 Explanations included:

Evolutionary Senility
Could not eat the evolving flowing plants
It got too cold
It got too hot
Mammals ate all of the eggs
They all died from a disease
They all became female when the climate changed
The sea level changed too fast
There was massive volcanism covering everything with ash
There was a nearby supernova of a star
Earth was hit by an asteroid


Alvarez, Alvarez, Asaro, and Mitchell All this changed in 1980 when Nobel laureate Luis Alvarez, his geologist son Walter Alvarez, nuclear chemist Frank Asaro, and paleontologist Helen Michael (on left, from right to left), published on their discovery of high levels of the element Iridium in a clay layer separating marine sediments of Cretaceous and Tertiary age.

Originally, Water Alvarez, a geologist at Berkeley, was looking for some way to quantify the rates of faunal change around the K-T boundary. To do this he needed a timekeeper. He and his father reasoned that the rain of dust from outer space should be coming down, on average, at a constant rate. Let's say it came down at a rate of 0.0001 g/yr on 1 cm2 of the ocean floor. It would be diluted by clay and microfossils also from the ocean to make sediment. So, if you had a 10 g sample of Cretaceous oceanic sediment with 1 g of space dust in it, it would have been deposited in 10,000 years [1 g / (0.0001 g/yr = 10,000 yr]. All you had to do was find a way of measuring space dust. It turns out that is not as hard as you might think.

 Remember, back in Lecture 10 that during the iron catastrophe most of the compliment of heavy elements that the Earth received during its accretion sank to its core. Iron was the main element, but along with it went most of the platinum and related elements called the platinum group. That way Platinum is a rare and expensive mineral on the Earth surface. Platinum group elements are thus much more common in your average space dust, than on the crust. These Platinum group elements would thus be a good signature for space dust in sediments. Better yet, Platinum group elements don't move around much once they are deposited. They are some of the so-called noble metals that tend to be unreactive and virtually inert. Of the Platinum group elements, Iridium (Ir) was relatively easy to measure, detectable in parts per billion. Thus Ir proved to be the element of choice to measure as a proxy for space dust.

K-T boundary, Gubbio, Italy The next thing to do was to find an outcrop of rock deposited in the deep ocean where the K-T boundary was well exposed and paleontologists were sure it was properly identified. Their choice turned out to be a highway cut in Gubbio, Umbria, Italy (on right). The boundary was clearly marked out by the disappearance of many kinds of Cretaceous microfossils, particularly most foraminiferans. At the boundary is a thin layer of brown and black clay. So the Alvarez group sampled carefully through this rock section and analyzed for Ir (below).

Iridum spike from Gubbio
(adapted from Alvarez, et al., 1980)

STEVEN KLINT, DENMARK They were astonished to find that there was very little change in Ir content through the section, except in the clay layer. They did their calculations only on the basis of leaching out all of the calcium carbonate from the samples, so they would only be measuring clay. Ir was up by more than an order of magnitude (factor of ten) in the clay bed, exactly where the extinctions occurred. They checked at two other sites, one in New Zealand and one at Stevns Klint in Denmark (right). What could have caused this? Two hypotheses were possible.

1) something shut off the production of clay, while the rain of Ir in space dust remained constant.

 2) something boosted the amount of space dust (and hence Ir) by and order of magnitude.

 They could not find a reason that it should be #1, so they opted for #2. But why should space dust go up? There could be two easy hypotheses to do this:

1) a nearby star could have gone supernova, showering the Earth with newly formed elements heavier than iron - among them Ir.

 2) the Ir could have come from a mass of extraterrestrial matter arriving in one chunk - as a giant meteor or comet.

Supernove 1987A in 1994 The key in science is not postulating causes, but rather testing hypotheses. So they reasoned that if it was # 1, one of the other elements that should have been created would be Plutonium (Pu), specifically Pu244, a radioactive isotope ("bomb" Plutonium). Pu244 was created during the supernova that led to our solar system (see Lecture 10), but nearly all has decayed to lead by now. However, if the Ir in the K-T boundary layer had come from a supernova 65 million years ago, there would be quite measurable amounts left. So they looked. The result was no measurable levels of Pu244. Their argument against a supernova was bomb proof (pun intended). However, it was a definite possibility. Click here to see links to a recent supernova.

Thus, they argued that this Ir-enriched layer was caused by the impact of a giant asteroid (~10 km) that put enough dust into the upper atmosphere to darken and hence cool the Earth for several years. This was theorized to result in shutting off global photosynthesis, with the resulting collapse of the global food chain. As a result nothing larger than 25 kg survived the boundary. This concept was promptly co-opted as a plausible scenario for the events following a nuclear holocaust (Nuclear Winter) as well. 


There were two big holes in the theory, however. One was that Ir enrichment could have occurred by volcanic eruptions as indicated elevated Ir levels in vapors from Kilauea Volcano in Hawaii. The other was that there was no impact site known.

A key observation was made by Bohor and Izett in 1986 when they discoed shocked quartz at Cretaceous-Tertiary boundary sections in the western US. Shocked quartz has only been observed in three settings:

1. High-pressure laboratory experiments.
2. Nuclear test sites.
3. Known giant meteor or asteroid impact sites.
Shocked quartz has never been found in any material from a volcanic eruption. This discovery, duplicated at may Cretaceous-Tertiary boundary sites, convinced many scientists that an impact was the only reasonable explanation of the boundary layer.

Oddly, in 1981, a year after the Alvarez et al. paper came out in Science, Glen Penfield and Antonio Camargo, petroleum geologists described what they interpreted to be a giant impact crater about the right age for the K-T boundary on the north shore of Yucatan in Mexico. Although they did publish their work, evidently they did not do it in a "flashy" enough journal, because their work was basically ignored until "rediscovered" by Alan Hildebrand and William Boynton in 1990. They were led search the literature for evidence in the Gulf of Mexico area by their discovery of thick layers of impact debris in Haiti and eastern Mexico that seemed to indicate that the impact site must have been quite near.

K-T boundary in Raton Basin Cretaceous-Tertiary boundary section in Raton Basin, New Mexico (left); shocked quartz (right). shocked quartz


Chicxulub crater, Yucatan Peninsula.
The impact structure, discovered buried beneath the shore of the Yucatan Peninsula (Chicxulub = "Mayan devil's tail"), turns out to be from 150 - 300 km in diameter, possibly the largest impact known in the world and said by some to be the largest known in the solar system. At the present diverse lines of evidence have largely confirmed the asteroid impact theory of mass extinctions and identified Chicxulub as the "smoking gun". Debate still rages by paleontologists about whether the impact was directly the cause of the extinction itself, or merely the coup de grace.


In the original Alvarez et. al. model, a 10 km bolide would have struck the earth sending a large amount of dust into the atmosphere. This would have blocked sunlight for a some period of time, resulting in massive cooling, globally, and a collapse of the food chain via. a cessation of photosynthesis.

 Since that time (1980) it has been realized that an oceanic impact, like that at Chicxulub could have much, dire effects.

In specific, the Chicxulub bolide struck a thick deposit of marine limestone (CaCO3) and underlying marine calcium sulphate (CaSO4 = Gypsum). This probably put large amounts of CO2 and sulfuric acid into the atmosphere within minutes. The CO2 would have produced an enhansed greenhouse effect, but the sulfuric acid would result in global cooling.
First, we will look at the greenhouse effect.
In 1989 O'Keefe and Aherns estimated that the impact of a 12 km diameter comet or a 14 km asteroid would have raised the average temperature of the Earth by over 10ƒ (right). This massive heating effect would have lasted for hundreds to thousands of years, probably making the interior of continents virtually inhospitable. Eventially the CO2 would be absorbed by the oceans, plants and the wweathering process. IMPACT

 Possible impact-generated CO2 effect on the atmosphere.


While there is now a general concensus amoung the geological and paleontological community that a giant impact did occur, there is still debate about its effect on the Earth's biota.

 This debate breaks down into two catagories.

1. Is the pattern of extinctions actually compatible with an impact.

 2. Were there other things going on that might have contributed to the extinctions, or even been more important. First we will look at the biological pattern.

Signor-Lipps Effect

Deccan Volcanism


 Possible CO2 effect on the atmosphere from Deccan Lavas. From Caldeira and Ramino (1990).

Note: A very nice chronology of the K-T boundaryu discoveries is given at:

VII: K-T Conclusions

  1. There was a giant impact of an asteroid or comet, producing an essentially global impact layer enriched in Ir and having shocked quartz, and various other impact related materials
  2. The impact site was Chicxulub.
  3. In oceans and on the continents the impact layer is at least coincidental with the mass extinction.
  4. On the continents a post-impact assemblage existed, dominated by ferns for hundreds to thousands of years.
  5. In Oceans, extinctions are directly followed by a large negative carbon isotopic shift suggesting a massive disruption of the biological carbon cycle: “Strangelove Ocean”.
  6. Fullerenes with extraterrestrial He isotopic ratios have been reported.
  7. There was massive outpourings of basaltic lava to form the Deccan Traps (~106 km3).
  8. K-T boundary is within Deccan Traps.
  9. Extinction of the dinosaurs opened up ecological spac allowing ascent of mammals.

VIII: Conclusions for the Triassic-Jurassic and Permian Triassic Boundaries


  1.  On Oceans and continents there was mass extinction at the Triassic-Jurassic boundary in both plants and animals.
  2. At least on the continents this extinction took place over less than 30 ky.
  3. As at the  K-T boundary, there is a “fern spike” indicating an assemblage dominated by ferns that lasted for hundred to thousands of years.
  4. There is a modest Ir anomaly at the boundary.
  5. Shocked quartz has been found in Italy.
  6. Manicouagan is a candidate impact of the right magnitude.
  7. But, the CAMP large igneous province is very close in time to the Triassic-Jurassic boundary. May be largest igneous event of all time (>106 km3).
  8. In oceans and continents extinctions are directly followed by a large negative carbon isotopic suggesting sudden and massive disruption of the carbon cycle.
  9. On continents there is evidence of a CO2 excursion based on stomatal density.
  10. Extinctions opened up ecological space allowing ascent of the dinosaurs to dominance with a consequent doubling of mass.


  1. On Oceans and continents there is mass extinction at the Permian-Triassic boundary in both plants and animals.
  2. In oceans and continents extinctions are directly followed by large negative delta 13C shift suggesting sudden and massive disruption of the carbon cycle.
  3. No “fern spike” has been found at the boundary.
  4. No Ir anomaly has been found at the boundary.
  5. No shocked quartz has been found.
  6.  Fullerenes with extraterrestrial He isotopic ratios have been reported.
  7. There are candidate impact sites.
  8. But Siberian Traps very close to boundary.
  9. Effects on the history of marine life massive, but effects less clear for continental life.

IX: Similarities between the Cretaceous-Tertiary, Triassic-Jurassic, and Permo-Triassic boundaries

  1. All three mass extinctions have evidence of extraterrestrial impacts: K-T undoubted, P-Tr, weakest (K-T: Ir, Shocked Quartz, fullerenes, known impact site; Tr-J: Ir, Shocked Quartz, plausible impact site; P-Tr: fullerenes, possible impact sites).
  2. Each of the three mass extinctions associated with one of the three largest continental flood basalt events.
  3. All three extinctions with large negative carbon isotopic shifts.
  4. All three extinctions associated with massive floral disruptions: two (K-T & Tr-J) with similar fern spikes.

Why the close parallelism?

  1. Culture ?
  2. Extinctions caused by LIPs (Large Igneous Provinces, such as the Deccan) ?
  3. Connection between impacts and volcanism ?

Caldeira, K. G. and Ramino, M. R., 1990, Deccan volcanism, greenhouse warming, and the Cretaceous/Tertiary boundary. Geological Society of America, Special Paper 247, p. 117-123.

 O'Keefe, J. D., and Aherns, T. J., 1989, Impact production of CO2 by the bolide and the resultant heating of the Earth. Nature, v. 338, p. 247-249.

Go to Lecture 25, The Cenozoic, the Evolution of Humans, and our Present Condition