GEOL 204 Dinosaurs, Early Humans, Ancestors & Evolution:
The Fossil Record of Vanished Worlds of the Prehistoric Past

Spring Semester 2021
Death from Below: The Permo-Triassic Mass Extinction

Detail from Julio Lacerda's reconstruction of the death of the Permian marine fauna.

"The way in which many Paleozoic life forms disappeared towards the end of the Permian Period brings to mind Joseph Hayden's Farwell Symphony where, during the last movement, one musician after the other takes his instrument and leaves the stage until, at the end, none is left." -- Curt Teichert, 1990

"The Mother of All Mass Extinctions" - Douglas Erwin, 1996

BIG QUESTION: What caused the Permo-Triassic mass extinction?

When Life Nearly Died

Permian/Triassic (251.902 Ma): The "Mother of All Mass Extinctions" (so named by Doug Erwin of the Smithsonian), this is the greatest diversity crisis known. If this was the single terminal Permian event, then it was an event with 55.7-82% of the marine genera went extinct (which corresponds to an 80-96% species level extinction). Or, to put it another way, there was only 4-12% survivorship at a species level, and given that a species could survive with very few individuals, it was much greater than 96% of individuals lost!). In comparison, the K/Pg had a 40-47% genus loss.

However, some models suggest that this is a two-phase extinction, with an earlier one between the Middle and Late Permian (the Guadalupian extinction), then EACH of these is among the greatest extinction events. As with the K/Pg there are many names for this event, including the Changhsingian/Induan extinction, or (using older terminology) the Tartarian/Scythian extinction.

This was a "game changing" event. Most estimates show that diversity of marine organisms was essentially stable from the Late Ordovician until the end of the Permian (minus the crash-and-recovery intervals at each mass extinction). In contrast, life ever since the Permian has been relatively steadily increasing in diversity, surpassing Paleozoic levels sometime in the Cretaceous.

Jack Sepkoski's studies for the evolution of marine life (which led to the recognition of the "Big 5" mass extinctions) did not merely look at general levels of diversity. He found that there were generally three different sets of organisms (not necessarily close relatives) who tended the share the same fates: when one member of each "evolutionary fauna" did well, the others were doing well, and when one suffered, they all suffered. He named these three evolutionary faunas the Cambrian, Paleozoic, and Modern faunas. (Don't let the names fool you! The "Cambrian fauna" is still present, but very rare; the Paleozoic fauna survives at moderate levels, and the Modern fauna goes back all the way to the Cambrian.)

The groups in each of the evolutionary faunas tend to share similar types of general traits:

The Permo/Triassic fauna wiped out most of the remaining Cambrian fauna, and was noticeably the time the
Paleozoic fauna stopped being the dominant assemblage and the Modern fauna took over.

The marine realm of the Permian was thus, like most of the Paleozoic, dominated by sessile epifaunal suspension feeders. There were nektonic predators around, and clams burrowing and snails crawling, but they were rarer than you would see in the seas today. On land the forests contained both conifers and other primitive seed plants and a great diversity of spore plants (horsetails, other ferns, club mosses, etc.): this is long before the rise of flowering, fruiting plants. The heyday of giant insects and millipedes and so forth was over, but there were still some of these around. Freshwater systems were patrolled by large amphibians, and the land dominated by the therapsid synapsids (protomammals), with reptiles (including the early precursors of the crocodilian-dinosaur group Archosauria showing up by the very end.)

In the marine realm, this was a major overhaul of diversity, with the world of the Mesozoic and Cenozoic radically different from that of the Paleozoic. Victims include:

There are major diversity losses in bryozoans (again), brachiopods (ditto), ammonoids (ditto), conodonts (ditto), those echinoderm groups that do survive, and terrestrial and marine vertebrates.

After the extinction, diversity was greatly reduced. Very, very few species were present, but some of these survivors were very common. We'll look more at patterns of survivorship in a bit.

Causes for the Greatest Extinction

Up until recently this was thought to be a very gradual event (see paleontologist Curt Teichert's quote at the top of these notes.). However, work in the 1990s established the catastrophic nature of this event. Some ideas were suggested as to what could cause such tremendous death at this time:

The ultimate cause appears to be the Siberian Traps, a huge lava field in Siberia with an area of 5 million km2, and a volume of about 3 million km3! (This would cover North America to a depth of 121 m (nearly 400'!). It erupted over the space of less than 1 Myr: a tremendous volume in a geologically-short interval. The Siberian Traps are the greatest example of a flood basalt: a form of eruption where vast volumes of lava (along with dissolved gases) spread out over huge regions in burst after burst. They are also called "large igneous provinces". The don't represent eruptions from a single conical volcano: rather, a huge region would have fissure eruptions.

The Earth Goes to Hell: Killing Agents of the Siberian Traps

However, although the Siberian Traps are the causal agent, the killing agent isn't the lava (except for the unlikely animals and plants of latest Permian Siberia!). Instead, the killing agents are produced mostly from the gases and aerosols released by the eruptions, and the side effects of these.

First and foremost is the carbon dioxide release, producing some of the most extreme global warming in Earth's history. Today (2018) there is about 873 Gt (gigatons) of C (carbon) in our atmosphere, reflecting about 410 parts per million CO2 (1 ppm = 2.13 Gt C). Before the Industrial Revolution, that value was around 600 Gt C. The Siberian Traps dwarfed these values by several orders of magnitude. Initial estimates were that over the course of the eruptions about 12,000-18,000 Gt C were released; newer modeling points to values closer to a mind-staggering 170,000 Gt C!! This didn't happen as a single burst, so there was no moment when Earth's atmosphere had 80,000 ppm CO2 during this interval, but it was still tremendously higher than today (or, more important for this issue, compared to the time before the extinction). Values aren't certain, but carbon dioxide levels had an increase of at least 8 times, and possibly more! This lead to extreme global warming (surface temperatures rising by more than 7°C, possibly much more), which lead to warming of the sea floor, which lead to melting of the methane clathrates (methane frozen in ice on the sea floor), which bubbled into the atmosphere, which led to even more global warming. The oceans would also become more acidic because of the carbon dioxide mixing with water, causing damage to shell-forming organisms.

Recent work points to the hyper-global warming as being first a jump from 710 to 2800 ppm, with a 38°C global average temperature. (For reference, in 2018 the global average temperature is 15.6°C [60.08°C].) After the full rise of CO2 and CH4 the estimates are that the atmosphere reached 5600 ppm, with global average 45°C (113°F)!!! This is hotter than a hot tub!

Additionally, the new studies point to as much as 18,000 Gt HCl (that's hydrochloric acid!) being released by the melting of underground rock salt deposits. Between the HCl, sulfuric acid (a byproduct of sulfur dioxide, a common component of eruptions), and carbon dioxide there would likely have been catastrophic levels of acid rain and oceanic acidification, worldwide.

Additionally, there were tremendous drops in atmospheric and oceanic O2 due to the mass death of so many land plants and phytoplankton and to oxidation of the methane. To make matters worse, increases in global temperature would decrease the temperature differential around the world, decreasing oceanic and atmospheric circulation, and thus reducing the churning up and oxygenation of the ocean water.

A consequence of the ocean anoxia is the promotion of sulfur bacteria. These prokaryotes prefer anoxic and dysoxic conditions. Their photosynthesis does not release oxygen as a waste product; instead, they release hydrogen sulfide (H2S), a gas highly poisonous to both aquatic and land animals. With shallow sunlit water devoid of their oxygen-generating competitors, the purple bacteria spewed out hydrogen sulfide, further making the shallow seas and land a hellish condition for other life.

But wait, there's more! The sulfates ejected by the eruptions and the H2S from the bacteria can destroy ozone. As these gases made their way into the higher atmosphere, the Earth lost its protection from dangerous UV radiation reaching the surface. Malformed pollen had already been known from latest Permian rocks; a recent experimental study showed that these malformations can be produced by subjecting plants of the same general groups present in the Permian with high levels of UV. Such malformations would greatly decrease plant fertility. (On top of that, the UV can kill the plants themselves, as well as animals.)

The sulfur aerosols did more than this, however. Their first effect was to block insolation, plunging the Earth into a temperature drop of 10-15°C for at least 10 years (some estimates suggesting a more profound extended 10s of kyrs cold snap.) But once the sulfate rains out of the atmosphere, it allowed the hyper-global warming to occur.

Stratigraphic evidence demonstrates that the main pulse of extinction took a mere 60 kyr, plus or mins 48 kyr, to occur!

And keep in mind, these killing agents are produced independent of living things. The eruptions would continue for 100s of kyrs, so that there would be long slow recovery for the atmosphere, ocean, and pedosphere (soil). This meant that the ecosystems continued to suffer for millions of years.

Patterns of P/Tr Extinction:

There are strong patterns to survivorship vs. extinction at the event:

So mollusks, arthropods, and vertebrates survive better than brachiopods, echinoderms, bryozoans, and cnidarians.

In the terrestrial realm, temperature stress, anoxia, hypercapnia (too much CO2), UV radiation, and probably acid rain were major killing agents. In the marine realm these plus sulfidication (okay, probably also affected the land!) and ocean acidification were at play.

In order to better test what was going on at the event, we need to be able to look at detailed stratigraphic ranges of fossils and long records of geochemical changes. In both cases, we need a record where we have rocks from before, at, and after the boundary. Thankfully, the American southwest, the Ural region of Russia, South Africa, and parts of China have all of these.

After the End

On land we see meandering streams temporarily disappearing (due to loss of ground-cover plants), and a great increase in the amount of fungal spores and hyphae in the fossil record (the decay of the rotting corpse of the Paleozoic Era.) The recovery fauna and flora was exceedingly depauperate (low in diversity): a small handful of taxa characterize both terrestrial and marine communities. Very recent work suggests a second round of extinctions a mere 180,000 years or so after the P/Tr boundary (although once again, we have to be concerned about stratigraphic resolution.)

In the seas aftermath of the P/Tr, there are many algal mats (stromatolites) in the shallow seas. Stromatolites are more characteristic of the world before animals; the rise of grazers like snails meant that stromatolites only form in conditions where animals can't live (hypersaline water, for instance). Also, as mentioned, most of the world's fauna and flora is made up of a few common species worldwide (rather than smaller numbers of individuals but greater diversity of species, with different species in different parts of the world.) There are no reefs and no calcareous algae in the seas in the Early Triassic; they only show up later in the Triassic. And there is some evidence that water temperatures were lethally hot at the equator: 40°C (104°F) or HIGHER!, rather than todays 25-30°C (77-86°F). Also, warm water holds less nutrients and less oxygen than cold water, making water not merely hot but starved. (Land temperatures would have also been phenomenally hot, but we cannot measure these as directly.)

On land, the best survivors among the amniotic vertebrates are: those that nest in burrows; those that may have been mountain-dwellers; and those which were semi-aquatic. All of these are groups which survive very well in low oxygen, high carbon dioxide conditions. Furthermore, groups of animals (like advanced protomammals and archosaurs) which very sophisticated methods of apparatus survive quite well.

Even though the main pulse of extinction occurred quickly, the Siberian Traps continued to erupt for hundreds thousands of years, keeping the planet's ecosystems destabilized throughout the Early Triassic. Only after they had settled down could Life get back to some form of normality. But the make up of the world had changed. The oceans became dominated by swimming and crawling and burrowing forms. And on land the Age of the Protomammals was over. The Age of Reptiles, and soon the Age of Dinosaurs, was at hand.

A "One-Two" Punch?

This is not the only major mass extinction associated with a flood basalt/large igneous province. The Triassic/Jurassic mass extinction is associated with the rifting of Pangaea and the eruption of the Central Atlantic Magmatic Province (CAMP). As we will see, the Cretaceous/Paleogene mass extinction has its Deccan Traps (although we will see there is something else also at play!). Some smaller extinctions also have LIPs associated with them: the Paleocene/Eocene and the North Atlantic event; the Early Jurassic/Middle Jurassic and the Karoo-Ferrar; etc.. V. Courtillot has shown there does seem to be a strong correlation, at least from the Early Permian onward. (Curiously, the mass extinctions seem to be linked to eruptions that followed long periods of magnetic stability, with a lag time of about 20 Myr after the "superchron" ends).

Recent evidence suggests that LIPs actually deliver a "one-two" punch. In both the P/Tr and Tr/J there is evidence for a short term (90 kyr or less) cold snap right at the beginning of each extinction. This might be due to the sulfates erupted into the atmosphere: stratospheric sulfates are long known to reflect a lot of sunlight, cooling the world beneath. Indeed, so much seems to be erupted at these two extinctions that there is evidence of a mini-duration but intense-level ice age, enough to cause a major drop of sea level and erosion (unconformities). (So maybe both Lyell and Cuvier were right: there is both a gap and a catastrophe at some extinctions!) The extended extreme volcanic winters themselves bring about extinctions, as might the crowding on the continental shelves as sea level drops. (A well-known aspect of ecology is that the area of a region is directly related to its species abundance: a rapid decrease of area produces a loss of species.)

Indeed, a "one-two" punch (or maybe in the case of the P/Tr a "one-two-three-four-five-six-seven-etc." punch!) might be necessary to produce one of the Big 5 mass extinction events?

Was my lecture on the Permo-Triassic Mass Extinction not enough for you? Here is another one, by Beniot Beauchamp (University of Calgary):

A relevant video:

To Lecture Schedule

Last modified: 10 January 2021

Life (such as it is) on the Siberian Traps: detail of figure of Julio Lacerda