The Permian–Triassic extinction event, also called the Great Dying, was a major extinction that happened at the end of the Permian period and the start of the Triassic period, marking the end of the Paleozoic era and the beginning of the Mesozoic era. It is the most severe extinction event in Earth's history, causing the loss of 57% of biological families, 62% of genera, 81% of marine species, and 70% of land vertebrate species. It also caused the greatest loss of insect species ever recorded. This event is the most extreme of the "Big Five" mass extinctions that occurred during the Phanerozoic eon. Scientists believe the extinction happened in one to three separate phases, with the largest marine extinction occurring over 60,000 to 100,000 years around 251.902 million years ago. Some debate exists about whether extinctions on land happened at the same time as the main marine extinction.
The main cause of the extinction was massive volcanic eruptions in an area called the Siberian Traps. These eruptions released large amounts of sulfur dioxide and carbon dioxide into the atmosphere. This led to oxygen-starved, sulfurous oceans, rising global temperatures, and acidified oceans. Atmospheric carbon dioxide levels increased from about 400 parts per million to 2,500 parts per million during this time, with an estimated 3,900 to 12,000 gigatonnes of carbon added to the ocean and atmosphere.
Other possible factors include the release of carbon dioxide from burning oil and coal deposits caused by the eruptions, methane emissions from the breakdown of methane clathrates, methane released by new types of microorganisms fed by minerals from the eruptions, stronger and longer El Niño events, and an extraterrestrial impact that formed the Araguainha crater. This impact may have caused the release of methane and damage to the ozone layer, increasing exposure to sunlight.
Dating
Previously, scientists believed that rock layers from the Permian–Triassic boundary were too few and incomplete to study in detail. However, new methods now allow scientists to date the extinction with precision down to thousands of years. Uranium-lead dating of zircon crystals from five volcanic ash layers at the Global Stratotype Section and Point in Meishan, China, has created a detailed timeline for the extinction. This timeline helps scientists explore how global environmental changes, carbon cycle disruptions, mass extinctions, and recovery processes occurred over thousands of years. The first appearance of the conodont Hindeodus parvus is used to mark the Permian–Triassic boundary.
The extinction occurred between 251.941 ± 0.037 and 251.880 ± 0.031 million years ago, lasting about 60,000 ± 48,000 years. A sudden drop in the ratio of carbon-13 to carbon-12 (δ¹³C) coincided with this event and is sometimes used to identify the Permian–Triassic boundary in rocks that cannot be dated using radiometric methods. This carbon isotope shift was 4–7% in magnitude and lasted about 500,000 years, though exact measurements are difficult due to changes in sediment layers over time.
Evidence suggests a global temperature increase of 8°C (14°F) and CO₂ levels rising to 2,500 ppm (compared to 280 ppm before the Industrial Revolution and 426 ppm today). Increased ultraviolet radiation reaching Earth also caused mutations in plant spores.
Some scientists believe the Permian–Triassic boundary is marked by a sudden increase in fungi, caused by the large amount of dead plants and animals available for them to feed on. This "fungal spike" has been used to identify the boundary in rocks unsuitable for radiometric dating or lacking index fossils. However, others argue that fungal spikes may have occurred multiple times after the extinction and that the most common fungal spore, Reduviasporonites, might actually be a type of algae. Recent chemical evidence supports a fungal origin for Reduviasporonites, reducing some criticisms.
Uncertainty remains about the total duration of the extinction and the timing of individual groups’ extinctions. Some studies suggest multiple extinction events, while others indicate a single peak. Research from Meishan, China, shows the main extinction clustered around one peak, but studies of other locations, such as Liangfengya and Shangsi, found evidence of two extinction waves with different causes. Paleoenvironmental analysis of rocks in Queensland, Australia, suggests repeated marine environmental stress before the end-Permian extinction, supporting the idea of a gradual process. Marine species diversity declined before the collapse of marine ecosystems, with a gap of about 61,000 years between these events.
Whether terrestrial and marine extinctions happened at the same time or at different times is still debated. Evidence from Greenland shows terrestrial and marine extinctions began together, with plants showing effects later. Many South China sediment layers show simultaneous extinctions, while other studies suggest terrestrial extinctions began 60,000 to 370,000 years before marine extinctions. Chemical analysis in Norway and dating of fossil layers in South Africa and China indicate terrestrial extinctions occurred before or after marine extinctions, depending on the region.
Studies of the Permian–Triassic extinction are complicated by the Capitanian extinction, a major event in the late Permian that may have occurred just before the Permian–Triassic extinction. Some extinctions once thought to occur at the Permian–Triassic boundary have been re-dated to the Capitanian period. It is unclear whether species that survived earlier extinctions were fully recovered before the Permian–Triassic event. Some scientists argue that environmental disasters over millions of years formed a single, long extinction event, while others suggest two major extinction pulses 9.4 million years apart. The first pulse, at the end of the Guadalupian epoch, caused the extinction of many species, including dinocephalian genera and the Verbeekinidae family of foraminifera. The impact of this event varied by location and species, with brachiopods and corals experiencing severe losses.
Extinction patterns
Marine invertebrates experienced the most severe losses during the P–Tr extinction. Earlier estimates of 90–96% marine species extinction were mixed with the end-Capitanian mass extinction, which happened 7–10 million years earlier. Evidence of these losses was found in rock layers from south China near the P–Tr boundary. In these layers, 286 out of 329 marine invertebrate genera disappeared within the last two sedimentary layers containing Permian conodont fossils. The drop in diversity likely happened because of a sudden rise in extinctions, not a decrease in new species forming.
This extinction mainly harmed organisms with calcium carbonate skeletons, especially those that needed stable CO₂ levels to build them. These organisms were harmed by ocean acidification caused by higher atmospheric CO₂. Organisms that used hemocyanin or hemoglobin to carry oxygen were more likely to survive than those using hemerythrin or oxygen diffusion. Endemism, or being found only in a specific area, was a strong risk factor. Bivalve species that lived in limited regions were more likely to go extinct than those found worldwide. Survival rates of species did not differ much by latitude. Organisms in areas less affected by global warming had fewer or delayed extinctions.
Among benthic (bottom-dwelling) organisms, the extinction increased background extinction rates, causing the greatest loss for species already prone to high background extinction rates. The extinction rate for marine organisms was extremely high. Bioturbators, organisms that mix sediments, were severely affected, as shown by the loss of mixed sediment layers in many marine environments during the end-Permian extinction.
Surviving marine invertebrates included articulate brachiopods (those with a hinge), which had been slowly declining since the P–Tr extinction; the Ceratitida order of ammonites; and crinoids ("sea lilies"), which nearly went extinct but later became abundant. Groups with high survival rates generally had strong control over circulation, complex gas exchange systems, and light calcification. Heavily calcified organisms with simpler breathing systems suffered the most. Among brachiopods, surviving species were usually small and rare compared to their former diverse communities.
Conodonts faced severe losses in both species and physical traits, though not as badly as during the Capitanian extinction. Ammonoids, which had been declining for 30 million years since the Roadian (middle Permian), experienced a major extinction event 10 million years before the P–Tr extinction, at the end of the Capitanian stage. This event greatly reduced the variety of ecological roles. Diversity and variety dropped further until the P–Tr boundary, where the extinction was non-selective, suggesting a sudden cause. During the Triassic, diversity rose quickly, but variety remained low. Ammonoid shapes became more limited as the Permian progressed. A few million years into the Triassic, their original range of forms was reoccupied, but the distribution of traits changed among groups.
Ostracods faced long-term diversity changes during the Changhsingian before the P–Tr extinction, when many suddenly disappeared. At least 74% of ostracods died during the P–Tr extinction itself. Bryozoans had been declining throughout the Late Permian and suffered even greater losses during the P–Tr extinction, being the most affected among lophophorates.
Deep-water sponges lost much diversity and had smaller spicules during the P–Tr extinction. Shallow-water sponges were less affected, showing larger spicules and less loss of physical diversity. Foraminifera faced a severe drop in diversity. Evidence from South China shows their extinction happened in two waves. Their diversity hotspots shifted to deeper waters during the P–Tr extinction. About 93% of late Permian foraminifera went extinct, including 50% of Textulariina, 92% of Lagenida, 96% of Fusulinida, and 100% of Miliolida. Foraminifera with calcium carbonate shells had a 91% extinction rate. Lagenida survived more than Fusulinida, possibly because they had greater environmental tolerance and wider geographic ranges.
Cladodontomorph sharks likely survived in deep ocean refugia, as shown by their discovery in deep, outer shelf environments during the Early Cretaceous. Ichthyosaurs, which evolved just before the P–Tr extinction, also survived.
The Lilliput effect, where species shrink during and after mass extinctions, was observed near the Permian-Triassic boundary in foraminifera, brachiopods, bivalves, and ostracods. While surviving gastropods were smaller than those that died, it is unclear if the Lilliput effect applied to them. Some gastropod groups, called "Gulliver gastropods," grew larger after the extinction, showing the opposite of the Lilliput effect, known as the Brobdingnag effect.
The Permian had many insect and invertebrate species, including the largest insects ever. The end-Permian extinction was the largest known for insects, with eight to nine orders going extinct and ten others losing diversity. Palaeodictyopteroids (insects with piercing mouthparts) declined during the mid-Permian, linked to changes in plant life. The worst decline happened in the Late Permian, likely not directly caused by weather-related plant changes. Some insect declines were due to biogeographic shifts rather than outright extinctions.
The geological record of terrestrial plants is limited, mostly based on pollen and spores. Plant changes across the Permian-Triassic boundary vary by location and preservation. Plants are generally not strongly affected by mass extinctions, with impacts at the family level being minor. Plant diversity losses were less severe than marine animal losses. A 50% drop in species diversity may be due to taphonomic processes (preservation biases). However, ecosystems changed greatly, with forests nearly disappearing and dominant plant groups like Cordaites (gymnosperms) and Glossopteris (seed ferns) declining rapidly. The extent of plant extinction remains debated.
The Glossopteris-dominated forests in high-latitude Gondwana collapsed in Australia about 370,000 years before the P–Tr boundary. This collapse was less severe in western Gondwana.
Biotic recovery
After the extinction event, the ecosystems changed as surviving species adapted and new ones appeared. In the oceans, ancient groups of animals declined, while modern groups became more common. The Permian-Triassic extinction was a major turning point in this shift, which began after an earlier extinction and ended in the Late Jurassic. Typical ocean floor animals included bivalves, snails, sea urchins, and crustaceans, while bony fish and marine reptiles became more common in open water. On land, dinosaurs and mammals first appeared during the Triassic. The change in which species were present was partly because the extinction event affected some groups more than others, like brachiopods compared to bivalves. However, recovery also varied: some species that survived later went extinct without returning to their former diversity, while others became dominant over time.
Right after the Permian extinction, ocean life was simple, with few species. A few types of animals, such as certain bivalves, conodonts, brachiopods, and foraminifera, dominated. These groups had low diversity, and the variety of species did not change much with latitude.
Scientists disagree about how quickly life recovered. Some say it took 10 million years until the Middle Triassic because the extinction was so severe. However, in some areas like Bear Lake County, Idaho, and parts of Nevada and China, life bounced back faster. For example, marine life in the Paris biota recovered in about 1.3 million years, and complex ecosystems in Italy and China showed signs of recovery less than a million years after the extinction. These differences suggest that some regions were less affected by the extinction, possibly because of better environmental conditions. High-latitude areas may have recovered faster due to higher productivity. While species diversity returned quickly, it took much longer for ecosystems to function as they did before the extinction. One study found that marine ecosystems were still recovering 50 million years after the event, even though species diversity had returned in a much shorter time.
Recovery speed also varied based on the type of organism. Seafloor communities remained simple until the end of the Early Triassic, about 4 million years after the extinction. Animals living on the ocean floor took longer to recover than those that burrowed into the seafloor. In contrast, swimming animals like ammonoids and conodonts recovered quickly, with ammonoids reaching pre-extinction diversity levels within 2 million years.
Recent research suggests recovery was driven by competition between species. In the Early Triassic, low competition slowed recovery, but as more species returned, competition increased, speeding up recovery. This created a cycle where more biodiversity led to more competition, which in turn led to even more biodiversity. However, some areas had delayed recovery because of repeated environmental problems, such as extreme heat and low oxygen levels. These conditions limited how complex ecosystems could become until the Spathian period.
By the Middle Triassic, most marine communities had fully recovered. However, global marine diversity reached pre-extinction levels only by the Middle Jurassic, about 75 million years after the extinction.
Before the extinction, about two-thirds of marine animals were attached to the seafloor. During the Mesozoic era, only about half of marine animals were attached, with the rest being free-moving. Fossil evidence shows fewer attached filter-feeding animals like brachiopods and sea lilies, and more mobile species like snails, sea urchins, and crabs. Before the Permian extinction, simple and complex marine ecosystems were equally common. After recovery, complex ecosystems outnumbered simple ones by about three to one. Increased predation and the ability of some animals to crush shells led to the Mesozoic Marine Revolution.
Marine vertebrates, like fish, recovered quickly, showing complex predator-prey relationships. Fossilized waste from 5 million years after the extinction suggests fish were already at the top of the food chain. Some fish species, like hybodonts, replaced teeth very quickly, and ichthyopterygians grew rapidly in size.
Bivalves quickly returned to many marine environments after the extinction. Before the Permian-Triassic extinction, bivalves were rare, but they became common and diverse in the Triassic, taking over roles that brachiopods had filled before the extinction. Some scientists once thought bivalves outcompeted brachiopods, but this idea is not fully supported by evidence.
Hypotheses about cause
Explaining events from 250 million years ago is very difficult because much of the evidence on land has been worn away or buried deep underground. The ocean floor spreads apart over time, and the old seafloor is completely reused over 200 million years, leaving no useful clues beneath the ocean.
Scientists have gathered strong evidence about the causes of this event, and several possible explanations have been suggested. These include both sudden, large-scale changes and slower, long-term processes, similar to those thought to cause the Cretaceous–Paleogene extinction event. However, scientists are not as certain about the causes of this event compared to that one.
Any explanation for the event must address why it affected certain organisms most, such as those with calcium carbonate skeletons. It must also explain why it took 4 to 6 million years before life began to recover, and why there was very little formation of biological minerals even though inorganic carbonates were deposited during recovery.
The flood basalt eruptions that created the Siberian Traps were among the largest volcanic events in Earth’s history. These eruptions covered an area about the size of Saudi Arabia (2,000,000 square kilometers) with lava. The timing of these eruptions matches the extinction event. Studies of the Norilsk and Maymecha-Kotuy regions in northern Siberia show that the volcanic activity happened in a few large bursts of magma, not in regular, steady flows.
The Siberian Traps caused one of the fastest increases in atmospheric carbon dioxide levels in Earth’s history. Scientists estimate that carbon dioxide emissions during this event were five times faster than during a previous mass extinction linked to the Emeishan Traps. Some studies suggest carbon dioxide levels rose from 500 to 4,000 parts per million before the extinction to about 8,000 parts per million afterward. Another study estimates that carbon dioxide levels were around 400 parts per million before the extinction and rose to 2,500 parts per million. This increase in carbon dioxide likely caused extreme global warming. Some evidence suggests a delay of 12,000 to 128,000 years between the rise in carbon dioxide and global warming, though this may be due to errors in dating.
Before the extinction, global average surface temperatures were about 18.2°C. These temperatures rose to as high as 35°C, with this extreme heat lasting up to 500,000 years. In the southern latitudes of Gondwana, temperatures increased by about 10–14°C. In South China, surface water temperatures near the equator rose by about 8°C. In present-day Iran, tropical sea temperatures increased from 27–33°C to over 35°C during the extinction. These temperature changes likely caused stronger El Niño events, making short-term climate changes more extreme.
High levels of carbon dioxide lasted for a long time because the arrangement of the supercontinent Pangaea made the carbon cycle less effective at removing carbon from the atmosphere. A 2020 study used a scientific model to show how the greenhouse effect from volcanic carbon dioxide emissions affected the ocean and led to the mass extinction. Evidence also suggests that volcanic activity burned underground fossil fuel deposits, as shown by spikes in mercury and carbon isotopes. Other signs, such as a 20-fold increase in certain chemical ratios, support that extreme volcanism occurred at the same time as the extinction. A large amount of isotopically light zinc from the Siberian Traps also confirms that volcanism was active during the extinction.
The Siberian Traps had unique features that made them especially dangerous. The region’s rocks contain high amounts of halogens, which can destroy the ozone layer. Evidence suggests that up to 70% of these halogens were released into the atmosphere, producing 18 teratonnes of hydrochloric acid and sulfur-rich gases. These gases created dust clouds and acid rain, blocking sunlight and harming photosynthesis in plants and ocean life. This would have disrupted food chains. However, these cooling effects were short-lived and unlikely to have caused widespread death.
Volcanic eruptions may also have caused acid rain, which could have harmed land plants and marine organisms with calcium carbonate shells. Flood basalts typically produce smooth, runny lava that does not send debris into the air. However, 20% of the Siberian Traps’ eruptions produced pyroclastic ash, which rose high into the atmosphere, increasing short-term cooling. Once this ash settled, carbon dioxide levels would have remained high, leading to long-term global warming.
The burning of hydrocarbon deposits may have worsened the extinction. The Siberian Traps are located above layers of ancient carbonate, evaporite, and coal deposits. When heated by volcanic activity, these rocks released large amounts of greenhouse gases and toxic fumes. The unique position of the Siberian Traps over these deposits likely made the extinction more severe. Volcanic eruptions that intruded into coal beds may have released even more carbon dioxide, intensifying global warming after the dust and ash settled.
A 2011 study found evidence that volcanic activity caused massive coal deposits to ignite, possibly releasing over 3 trillion tons of carbon. Ash deposits near the Buchanan Lake Formation suggest that toxic elements were released into water bodies. However, some scientists argue that these ash layers may have come from wildfires unrelated to volcanic activity. A 2013 study estimated that the Siberian Traps released large amounts of carbon dioxide, carbon monoxide, hydrogen sulfide, and sulfur dioxide.
The way the Siberian Traps erupted, with magma forming thick layers (sills) rather than spreading as lava, may have prolonged their warming effects.
Comparison to present global warming
The Permian-Triassic mass extinction (PTME) has been compared to today’s human-caused global warming and the Holocene extinction because all three involve fast increases in carbon dioxide levels. While current greenhouse gas emissions are more than ten times faster than those during the PTME, the timing and pattern of carbon release during the PTME are not fully understood. Scientists believe the PTME’s carbon release likely happened in short, intense bursts over a few key periods, not continuously. During these bursts, the rate of carbon release may have been similar to today’s human-caused emissions.
Today’s oceans are also experiencing lower pH and oxygen levels, similar to what happened during the PTME. This connection has led scientists to compare modern ecosystems with those of the PTME. If carbon dioxide levels keep rising, another event similar to the PTME’s biocalcification crisis—where marine life that builds calcium carbonate structures is severely affected—may occur. Changes in how plants and insects interact during the PTME have also been studied as possible signs of future ecological changes.
Geologists warn that the similarities between the PTME and today’s situation show the urgent need to reduce carbon dioxide emissions to avoid a similar disaster.
As during the PTME, today’s oceans are changing rapidly, with falling pH and oxygen levels. This strengthens the link between the two events. Geologist Lee Kump explains:
The Permian-Triassic mass extinction shows the serious effects of rapid carbon dioxide emissions. Volcanic activity during the PTME released huge amounts of CO₂, causing ocean acidification, low oxygen levels, and major ecological damage. Today, human activities are causing similar changes, but even faster. The geological record shows that once ecosystems reach certain tipping points, the effects can last for millions of years.
If carbon dioxide levels continue to rise, another biocalcification crisis—like the one seen in the fossil record—could happen. This would have severe consequences for modern marine ecosystems.