Permian–Triassic Extinction Event

The Permian–Triassic boundary is formally defined at the GSSP at Meishan, Zhejiang Province, China, by the first appearance datum (FAD) of the conodont Hindeodus parvus, dated to 251.902 ± 0.024 Ma. The high-precision U-Pb zircon geochronology of Burgess, Bowring, and Shen (2014), using the EARTHTIME tracer calibration, established that the main extinction interval spanned from Bed 25 (251.941 ± 0.037 Ma) to Bed 28 (251.880 ± 0.031 Ma) at Meishan, giving a maximum duration of 61 ± 48 thousand years. This revised estimate is substantially shorter than earlier estimates of up to 1.5 million years, and is broadly consistent with astrochronological estimates of approximately 83 thousand years.

Detailed study of South Chinese boundary sections has resolved the crisis into a two-phase event. The first extinction pulse occurred at the top of Bed 24 at Meishan, at the transition between the Clarkina yini and C. meishanensis conodont zones, eliminating approximately 60% of species regionally. Losses were particularly severe among calcareous algae, fusulinid foraminifers, gastropods, and ammonoids. A second pulse occurred in the earliest Triassic Isarcica staeschi Zone, around the top of Bed 28, causing approximately 70% species-level losses with especially heavy casualties among ostracods, brachiopods, and small foraminifers. The interlude between the two crisis phases lasted approximately 55 thousand years and is characterized by a mixed fauna of Permian survivors and newly originating taxa, along with the proliferation of microbialites and oolites in shallow-water settings.

The magnitude of the extinction is unmatched in the Phanerozoic. Estimated losses include approximately 57% of all biological families, 62% of genera, 81% of marine species, and 70% of terrestrial vertebrate species. Among marine invertebrates, suspension feeders and reef-dwelling organisms were particularly devastated. Several major groups with deep Paleozoic roots were permanently extinguished: trilobites, which had persisted for nearly 300 million years; rugose and tabulate corals that had constructed reefs since the Ordovician; fusulinid foraminifers; blastoid echinoderms; eurypterids; acanthodian fish; and placoderms (though the latter two groups were already in severe decline). The Paleozoic Evolutionary Fauna described by Sepkoski—dominated by brachiopods, crinoids, rugose corals, and bryozoans—was dismantled and replaced by the Modern Evolutionary Fauna dominated by bivalves, gastropods, and bony fishes.

On land, the dominant Southern Hemisphere Glossopteris flora was destroyed and replaced. Eight orders of insects became extinct, and approximately two-thirds of terrestrial tetrapod families were lost. Among the few surviving tetrapod lineages were procolophonoids, dicynodonts, therocephalians, cynodonts, and archosauromorphs. The boundary is also associated with a "coal gap" lasting approximately 6–10 million years, indicating a severe depletion of peat-forming plant communities, and a so-called "fungal spike"—an anomalous abundance of fungal spores and hyphae interpreted as evidence for massive terrestrial die-off and decomposition of dead plant matter.

The Siberian Traps Large Igneous Province (LIP) is widely accepted as the primary trigger for the extinction. This enormous volcanic province, covering more than 2 million km² of present-day Siberia, erupted over a period of approximately 1–2 million years bracketing the Permian–Triassic boundary. Burgess and Bowring (2015) and Burgess, Muirhead, and Bowring (2017) established a refined chronological framework dividing Siberian Traps magmatism into three stages. Stage 1, prior to the extinction, was characterized by initial pyroclastic eruptions followed by flood lava effusion, during which an estimated two-thirds of total lava volume (>1 × 10⁶ km³) was emplaced with relatively little biospheric disruption. Stage 2 began at 251.907 ± 0.067 Ma, marked by cessation of extrusive eruption and the onset of widespread sill-complex formation into the thick, volatile-rich Tunguska sedimentary basin. Stage 3, beginning at 251.483 ± 0.088 Ma, saw the resumption of lava extrusion alongside continued sill intrusion.

The critical insight is that the onset of Stage 2—the initial pulse of sill emplacement—coincides precisely with the onset of both the mass extinction and the abrupt negative δ¹³C excursion. Heat from sill intrusions into the Tunguska basin's evaporite, carbonate, clastic, and hydrocarbon-bearing sediments is estimated to have generated massive volumes of greenhouse gases (CO₂, CH₄) and other volatiles through contact metamorphism. The fact that voluminous lava eruptions during Stage 1 produced minimal environmental disturbance, while the shift to intrusive magmatism triggered catastrophic change, demonstrates that it was not the sheer volume of magma but rather its interaction with volatile-rich sediments that proved lethal. This model has broader implications: LIPs characterized by extensive sill complexes emplaced into volatile-fertile sedimentary basins are more likely to trigger mass extinctions than those dominated by flood basalts and dikes.

The end-Permian extinction was driven by a cascade of interconnected environmental stresses, most of which stem from the massive release of CO₂ and other greenhouse gases.

Global warming: Conodont apatite δ¹⁸O records indicate that sea surface temperatures rose by approximately 10°C during the extinction interval, from roughly 23°C to 33°C, and possibly exceeded 35–40°C in the Early Triassic tropics. Such lethal temperatures in shallow tropical waters may explain the selective loss of surface-dwelling groups and the migration of some surviving foraminifers to deeper, cooler habitats.

Ocean anoxia and euxinia: Sedimentological, petrographic, and geochemical evidence—including pyrite framboid size populations, uranium isotope ratios (δ²³⁸U), iron speciation, and trace metal distributions—indicates that ocean anoxia expanded dramatically during the extinction, developing across a range of settings from continental shelves to abyssal depths. In many regions, conditions progressed beyond simple anoxia to euxinia (free hydrogen sulfide in the water column), particularly in high-latitude Boreal settings. Earth-system models suggest that warming-induced increase in freshwater runoff to northern oceans created density stratification that weakened deep-water formation, resulting in poorly ventilated global oceans.

Carbon cycle disruption: The extinction is associated with a pronounced negative δ¹³C excursion of approximately 3–6‰, reflecting massive injection of isotopically light carbon into the ocean-atmosphere system. Carbon cycle volatility persisted for approximately 500 thousand years before values partially recovered, and continued perturbation is recorded well into the Early and Middle Triassic.

Other proposed mechanisms: Ocean acidification has been proposed based on calcium isotope anomalies and boron isotope data, but evidence remains conflicting. Ozone layer destruction by volcanic halogens and hydrogen sulfide released from euxinic oceans has been invoked to explain land-based extinction. Mercury toxicity from volcanic emissions has also been suggested as a contributing factor. Most researchers favor a synergistic model in which multiple kill mechanisms operated simultaneously, with warming, anoxia, and hypercapnia (elevated CO₂) acting together.

The aftermath of the extinction was characterized by dramatically simplified ecosystems, the proliferation of disaster taxa, and a protracted recovery that is among the longest known for any mass extinction.

In the marine realm, the immediate post-extinction environment was dominated by microbialites, oolites, and low-diversity assemblages of opportunistic species. The "chert gap" in oceanic sediments reflects the near-complete loss of radiolarian productivity. Planktonic communities shifted from eukaryotic algae to assemblages dominated by cyanobacteria, green-sulfur bacteria, and prasinophytes. Nitrogen isotope records (δ¹⁵N) dropping to ~0‰ indicate that nitrogen-fixing cyanobacteria became dominant, suggesting severe nutrient limitation in surface waters.

On land, the dicynodont Lystrosaurus has long been cited as the quintessential disaster taxon, reportedly comprising up to 90% of Early Triassic terrestrial vertebrate individuals. The vegetation collapse caused a shift from meandering to braided river systems across multiple continents, as the loss of root-binding vegetation destabilized soils. The coal gap, lasting through the Early and into the Middle Triassic, signals the absence of forests capable of producing peat.

The pace of recovery varied dramatically between groups. Nektonic organisms such as ammonoids showed relatively rapid taxonomic recovery, with quick diversification in the Early Triassic, though these early radiations were themselves repeatedly disrupted. Benthic marine communities, however, displayed an extended Early Triassic lag phase lasting approximately 5 million years, during which diversity remained very low. The main stage of benthic recovery did not commence until the middle Anisian (early Middle Triassic, ~247 Ma), when a hyperbolic increase in species richness began, driven largely by the resurgence of carbonate platforms and the intensification of biotic interactions. Gastropods in particular diversified explosively in carbonate platform settings, likely in association with the proliferation of dasycladacean algae.

For terrestrial vertebrates, the picture is even more protracted. Although global familial diversity nominally recovered by the Olenekian (~250–245 Ma), this reflects the filling of ecological space by disaster taxa rather than true ecosystem maturation. Community-level (alpha) diversity never recovered to pre-extinction Artinskian levels during the Middle Triassic. Guild structure, body-size distributions, and trophic complexity were slowly rebuilt, with the final restoration of fully complex terrestrial ecosystems not achieved until the Carnian stage (~237–227 Ma), approximately 30 million years after the extinction. It was only in the Late Triassic, with the radiation of dinosaurs, pterosaurs, crocodilians, mammals, and other groups, that terrestrial biodiversity was fully restored.

The end-Permian extinction did not occur in isolation. It was preceded by at least two earlier extinction pulses during the Guadalupian epoch. A significant event at the end of the Guadalupian (~260 Ma), sometimes called the end-Capitanian extinction, destroyed substantial marine diversity and may have been associated with the Emeishan LIP volcanism in South China. On land, an even earlier event at the Kungurian-Roadian boundary (~271 Ma), termed "Olson's Gap" or "Olson's Extinction," saw approximately two-thirds of terrestrial vertebrate diversity lost. These earlier events weakened ecosystems that had only partially recovered by the time the terminal Permian extinction struck, compounding the devastation.

The Late Permian world was characterized by the supercontinent Pangaea, which brought most of Earth's landmasses together. This configuration reduced the total area of continental shelves—the most productive marine habitats—and created vast continental interiors prone to extreme climate. The single global ocean, Panthalassa, and the enclosed Tethys seaway meant that disruptions to ocean circulation could propagate globally with little buffering. The assembly of Pangaea itself had already reduced marine biodiversity by eliminating separate biogeographic provinces, so the biosphere was arguably in a vulnerable state when the Siberian Traps volcanism began.

The Permian–Triassic extinction is the defining boundary between the Paleozoic and Mesozoic eras and represents the most fundamental restructuring of global ecosystems in the Phanerozoic. It permanently ended the dominance of the Paleozoic Evolutionary Fauna and set the stage for the Mesozoic world, ultimately enabling the radiation of groups that would dominate subsequent eras—including dinosaurs, modern corals (scleractinians), modern bony fishes, and eventually mammals. All modern sea urchins, for example, descend from just two Paleozoic genera that survived the extinction. The event also serves as a sobering natural analog for understanding the potential consequences of rapid anthropogenic carbon emissions and global warming, as emphasized by Penn et al. (2018) and other recent studies. The speed of extinction (tens of thousands of years), the role of greenhouse gas-driven warming and ocean deoxygenation, and the extraordinary duration of recovery all carry direct relevance for understanding the potential trajectories of the current biodiversity crisis.