{
  "id": "mass-extinction",
  "title": "Mass Extinction",
  "lang": "en",
  "source_lang": "en",
  "summary": "A mass extinction is a geologically rapid, global loss of biodiversity that eliminates an unusually large proportion of species across many groups.",
  "category": "science",
  "subcategory": "earth-science",
  "type": "concept",
  "aliases": [
    "Mass-extinction event",
    "biotic crisis",
    "extinction event"
  ],
  "generated_at": "2026-11-03",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/mass-extinction/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/mass-extinction/",
      "json": "https://aiwiki.page/zh/mass-extinction.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "A mass extinction is an episode in which the global rate of extinction rises far above its usual level, eliminating a substantial proportion of biodiversity across many evolutionary lineages and environments. Unlike the disappearance of a single species or a regional community, it represents a disruption of the biosphere on a planetary scale. “Rapid” is understood geologically: an extinction pulse may last thousands of years, whereas a broader crisis can contain repeated pulses separated by hundreds of thousands or millions of years.\n\nFive especially severe crises are conventionally recognized during the last approximately 539 million years: the end-Ordovician, Late Devonian, end-Permian, end-Triassic, and end-Cretaceous extinctions. These “Big Five” differed in duration, causes, and ecological consequences. Their study connects paleontology, geology, and evolution, showing how environmental disruption can overwhelm biological systems and redirect the subsequent history of life. Comparisons with present-day biodiversity loss are important, but require careful distinctions between measured extinctions, population declines, threatened species, and projected future losses."
    },
    {
      "heading": "Definition, magnitude, and the Big Five",
      "text": "Extinction is a normal component of evolution. Species originate and disappear as environments change, competitors spread, and populations become unable to persist. This continuing turnover is often called background extinction. A mass extinction differs primarily in its exceptional intensity, geographic extent, and breadth across groups. It need not kill all organisms, affect every environment equally, or eliminate the same proportion of species in every lineage. An event can be globally catastrophic while leaving some ecosystems and taxonomic groups comparatively intact.\n\nA commonly used benchmark is the loss of at least 75 percent of species over a geologically short interval. This is a practical convention rather than a universal biological threshold. The fossil record rarely permits a direct census of all species living before and after a crisis, and extinction estimates depend on geographic coverage, time resolution, and the taxonomic level counted. Losses of families or genera cannot be translated into species losses by a fixed multiplier. Estimates such as “three-quarters of species” should therefore be understood as reconstructions with uncertainty, not exact historical inventories.\n\nThe Big Five framework became influential following David Raup and J. John Sepkoski Jr.’s 1982 analysis of marine family-level extinction through the Phanerozoic Eon. Their work identified unusually severe intervals against a changing background rate. Later compilations, statistical methods, and improved dating refined the pattern. The approximate dates are 444 million years ago for the end-Ordovician crisis, about 372 and 359 million years ago for major Late Devonian crises, 252 million years ago for the end-Permian event, 201 million years ago for the end-Triassic event, and 66 million years ago for the end-Cretaceous event.\n\nThese five are not a complete list of biologically important crises. The mid-Permian Capitanian extinction, for example, caused substantial losses but is not ordinarily included in the traditional Big Five. Earlier Precambrian biological transitions are harder to compare because organisms, preservation, and available taxonomic information differ greatly from those of later eras. Nor must every mass extinction form a sharply separate statistical class: some analyses describe a continuum extending from ordinary turnover to extreme events.\n\nThe defining feature is elevated extinction, not merely a reduction in net diversity. Diversity also depends on speciation. A long interval with little origination can produce a substantial diversity decline without an exceptionally intense killing episode. This distinction is particularly relevant to prolonged crises such as the Late Devonian, in which repeated extinction and reduced origination both contributed to biological impoverishment."
    },
    {
      "heading": "Reconstructing extinction from rocks and fossils",
      "text": "The main evidence comes from fossils arranged within successions of sedimentary rocks. Researchers compare the last appearances of organisms across many locations, seeking losses that coincide despite differences in local environment. Marine animals with mineralized shells dominate many datasets because they preserve relatively well and occur in extensive sedimentary deposits. The history of terrestrial organisms, soft-bodied animals, and many microorganisms is less completely represented. A global extinction curve is consequently a reconstruction from uneven evidence rather than a direct record of the entire biosphere.\n\nA central difficulty is the Signor–Lipps effect: incomplete sampling generally places a species’ last observed fossil below its true disappearance. A sudden extinction can therefore appear gradual. The opposite problem also occurs. Older fossils can be eroded from sediment and redeposited in younger layers, creating an apparent survival after extinction. Changes in sea level or sedimentation can remove particular habitats from a local rock succession, producing disappearances that reflect migration or missing deposits rather than global extinction.\n\nResearchers address these problems through geographically extensive sampling, explicit treatment of preservation, and statistical comparisons among deposits. Occurrence databases record where and when taxa have been found, allowing analyses that account for unequal collecting effort. Methods such as coverage-based standardization reduce some differences between heavily and sparsely sampled intervals. Nevertheless, the result depends on taxonomic consistency, the environmental range represented, and the time span assigned to each sample. A million-year bin can conceal several distinct pulses.\n\nDating combines fossil correlations with physical and chemical markers. Radiometric dating, particularly uranium–lead measurements of volcanic zircon, can constrain the ages of ash beds associated with fossil-bearing strata. Magnetic reversals, sedimentary cycles, and changes in isotope ratios help connect distant sections. Carbon-isotope excursions can identify disturbances of the carbon cycle, although a shared excursion does not by itself prove that all associated biological changes occurred simultaneously.\n\nThe interpretation of extinction changed substantially during the history of geology. Georges Cuvier’s work on fossil vertebrates in the late eighteenth and early nineteenth centuries helped establish that extinction was real. Later research increasingly emphasized long-term environmental change and ordinary geological processes. In 1980, Luis Alvarez, Walter Alvarez, Frank Asaro, and Helen Michel proposed an extraterrestrial impact as the cause of the end-Cretaceous extinction, based partly on an iridium anomaly at the boundary. Subsequent identification of the Chicxulub crater and its age transformed this hypothesis into a well-supported explanation. More recent work has increasingly combined fossil records, high-precision dating, geochemistry, and numerical models to distinguish immediate killing mechanisms from longer environmental deterioration."
    },
    {
      "heading": "The end-Ordovician and Late Devonian crises",
      "text": "The end-Ordovician extinction occurred about 444 million years ago, when most well-documented animal diversity inhabited the sea. It is usually reconstructed as two major pulses associated with large climatic and sea-level changes. Estimates commonly place total marine species losses around 85 percent, although the precise figure remains uncertain. Brachiopods, bryozoans, trilobites, graptolites, and other marine groups suffered extensive losses. Because complex terrestrial ecosystems were still limited, the event is known principally from marine evidence.\n\nAn important driver was the growth of a major ice sheet on Gondwana, which then extended across the South Pole. Cooling altered climatic zones, while falling sea level reduced the shallow continental seas that supported diverse communities. Species adapted to warm water or restricted habitats faced both thermal stress and habitat contraction. The later phase of the crisis accompanied deglaciation, renewed sea-level rise, and changes in ocean circulation and oxygenation. The details vary among regions, but oxygen-poor conditions were important in many marine settings.\n\nThis combination illustrates why a mass extinction cannot always be assigned to a single environmental variable. Cooling may initially eliminate warm-adapted species, while subsequent warming and oxygen loss threaten the survivors. A rapid sequence of opposing changes can be especially destructive because organisms that persist through one phase are not necessarily suited to the next. The causes of the initial Ordovician cooling and the exact relationship between glaciation and marine oxygen depletion remain active research questions.\n\nThe Late Devonian crisis was more prolonged. Its best-known episodes include the Kellwasser events near the Frasnian–Famennian boundary, approximately 372 million years ago, and the Hangenberg event near the Devonian–Carboniferous boundary, approximately 359 million years ago. Reef ecosystems suffered severe damage, including major losses among stromatoporoid sponges and corals. Various brachiopods, trilobites, ammonoids, and fishes also declined. The Hangenberg event affected both marine and freshwater vertebrate communities, including major changes among armored fishes and early limbed vertebrates.\n\nWidespread deposition of organic-rich sediments provides evidence for repeated ocean anoxia, meaning severe depletion or absence of dissolved oxygen. Several mechanisms may have contributed, including climatic fluctuations, sea-level changes, volcanism, and nutrient delivery from land. The expansion of deeply rooted forests could have enhanced rock weathering and nutrient transport to the sea, stimulating productivity and subsequent oxygen consumption as organic matter decayed. This is a plausible contributing mechanism rather than a complete, universally accepted explanation.\n\nThe Devonian record is especially important because it challenges the image of every mass extinction as a single brief catastrophe. Repeated environmental shocks, regional differences, and intervals of suppressed origination produced a long restructuring of ecosystems. Different groups disappeared at different stages, and the causes and duration of one pulse cannot automatically be assigned to the entire crisis."
    },
    {
      "heading": "The end-Permian extinction",
      "text": "The end-Permian extinction, approximately 252 million years ago, was the most severe of the Big Five. It marks the boundary between the Paleozoic and Mesozoic eras. Modern estimates commonly indicate the loss of more than 80 percent of marine species, with some older reconstructions exceeding 90 percent. Exact percentages remain sensitive to methods and taxonomic assumptions. Terrestrial ecosystems also experienced major disruption, although land records are more geographically fragmented and their relationship to the marine extinction is not everywhere resolved at the same temporal precision.\n\nAmong the marine casualties were the last trilobites, rugose and tabulate corals, and many brachiopods, crinoids, and other bottom-dwelling organisms. Some once-prominent groups survived only as reduced remnants. On land, changes affected plant communities, insects, and vertebrates, including the mammal-line reptiles known as therapsids. The crisis did not simply remove rare species; it transformed the composition and functioning of entire communities.\n\nThe leading explanation links the extinction to the Siberian Traps, a vast episode of volcanic and intrusive activity. Lava eruptions released gases, while magma intruding into carbon-rich and evaporite-bearing sediments could generate additional climate-active and chemically reactive compounds. High-precision dating places important phases of this magmatism close to the extinction interval. The relevant environmental disturbance was therefore not confined to lava-covered areas: gases and their consequences propagated through the atmosphere and oceans.\n\nLarge releases of carbon dioxide contributed to rapid warming. Warmer seawater holds less oxygen, while higher temperatures increase the oxygen requirements of many animals. Changes in circulation and increased microbial respiration further expanded oxygen-poor waters. Some settings became euxinic, containing both little oxygen and dissolved hydrogen sulfide. Ocean acidification likely imposed additional stress, particularly on organisms that formed calcium carbonate skeletons. The relative contribution of each process differed among habitats and remains a subject of investigation.\n\nThe abrupt marine losses appear to have unfolded over a comparatively short interval, with some well-studied sections suggesting a duration on the order of tens of thousands of years. Such estimates describe particular extinction horizons and should not be equated with the entire period of environmental disruption. Repeated warming episodes and unstable marine chemistry continued into the Early Triassic.\n\nRecovery was correspondingly uneven. Some organisms increased rapidly in disturbed environments, but the rebuilding of diverse reefs, complex food webs, and many animal communities required millions of years. The end-Permian record demonstrates that survival through the initial extinction does not guarantee prompt ecological recovery: persistent environmental stress can constrain ecosystems long after the largest pulse of losses."
    },
    {
      "heading": "The end-Triassic extinction",
      "text": "The end-Triassic extinction occurred approximately 201 million years ago, shortly before the opening of the Atlantic Ocean progressed through the breakup of Pangaea. It eliminated numerous marine organisms and substantially reorganized terrestrial vertebrate communities. Conodonts, small eel-like vertebrates known mainly from their tooth-like feeding elements, disappeared entirely. Ammonoids suffered major losses, as did several groups of bivalves, brachiopods, and reef-building organisms. On land, many large non-dinosaurian archosaurs disappeared, including phytosaurs and aetosaurs.\n\nThe event is closely associated with the Central Atlantic Magmatic Province, one of the largest known continental igneous provinces. Its remnants occur around the modern Atlantic margins in North and South America, Europe, and Africa. High-precision dates show that major eruptive activity occurred near the extinction interval. As with the Siberian Traps, the important global effects came largely from gases and environmental change rather than direct contact with lava.\n\nVolcanic carbon dioxide increased greenhouse forcing, while sulfur emissions could produce shorter-lived cooling episodes. This difference in atmospheric residence time matters: repeated eruptions can alternate transient cooling with sustained warming. Fossil and geochemical evidence indicates major disturbances of the carbon cycle, with ocean warming, acidification, and oxygen depletion implicated in marine losses. Individual eruption pulses and biological changes are not equally well resolved in all regions, so the precise sequence continues to be refined.\n\nThe terrestrial record indicates strong ecological turnover, but responses were geographically variable. Some vegetation records show abrupt changes, increased fire activity, or disturbances in pollen and spore assemblages. These signals cannot automatically be treated as a uniform global disappearance of forests. The preservation of plant remains depends heavily on local sedimentary environments, and plant communities may change composition without generating species-level extinction patterns identical to those of animals.\n\nDinosaurs had originated well before the crisis and were already important in some Late Triassic ecosystems. Their subsequent expansion should therefore not be described as their sudden creation by the extinction. Instead, the removal of competing or ecologically similar groups contributed to opportunities for surviving dinosaur lineages during the Jurassic. Crocodile-line archosaurs also survived, despite the loss of many of their relatives.\n\nWhy particular lineages persisted is not fully resolved. Geographic distribution, physiology, body size, reproductive biology, and habitat use may all have mattered. Proposed advantages of particular dinosaur groups remain difficult to test independently because these traits occur together and are incompletely known in fossils. The event provides a clear case of volcanic activity associated with global biological disruption, while leaving substantial questions about the exact traits that determined survival."
    },
    {
      "heading": "The end-Cretaceous extinction",
      "text": "The end-Cretaceous extinction occurred 66 million years ago at the Cretaceous–Paleogene boundary. Approximately three-quarters of species are commonly estimated to have disappeared. The best-known casualties were non-avian dinosaurs, but the event also eliminated ammonites, pterosaurs, and several major groups of marine reptiles. Birds survived as one branch of dinosaurs, although many bird lineages also vanished. Marine plankton experienced severe losses, especially among some groups with calcium carbonate shells, disrupting food supplies across ocean ecosystems.\n\nThe principal trigger was the Chicxulub impact on the present-day Yucatán Peninsula in Mexico. The impactor was approximately 10–15 kilometres across, and the resulting crater is about 180 kilometres in diameter. Evidence includes a worldwide boundary layer enriched in iridium, shocked minerals, impact spherules, ejecta deposits, and the crater itself. The correspondence between the impact horizon and abrupt fossil changes provides unusually strong evidence for a specific event causing a mass extinction.\n\nThe impact released enormous mechanical and thermal energy, excavating carbonate- and sulfate-rich rocks. Dust, sulfur-bearing aerosols, and soot could reduce sunlight and cool the surface, producing an impact winter. The severity and duration of darkness depended on particle properties and atmospheric processes, which remain subjects of modeling. Suppression of photosynthesis would have reduced primary production and interrupted food webs. Additional effects included intense regional heating, fires, tsunamis, acid deposition, and rapid changes in ocean chemistry.\n\nThe Deccan Traps of India were also erupting around the boundary interval. Their emissions altered climate before and after the impact, and research continues into how volcanic and impact-related stresses interacted. The timing and magnitude of gas release are more important for this question than lava volume alone. Although volcanism contributed to environmental conditions, the impact is the best-supported primary cause of the abrupt boundary extinction.\n\nSurvival was selective rather than random. Freshwater ecosystems often fared better than marine and terrestrial systems dependent on continuous photosynthetic production, partly because detrital food chains could use stored organic matter. Small body size, flexible diets, burrowing, and access to sheltered habitats may have helped particular survivors, but no single trait explains every case. Sharks, crocodilians, turtles, mammals, and birds all survived as groups while losing some component lineages.\n\nThe extinction reshaped terrestrial ecosystems without instantly producing their later form. Mammals diversified after the boundary, but their expansion varied among regions and ecological roles. Surviving birds similarly underwent substantial evolutionary diversification. Recovery of marine productivity could occur faster than restoration of the previous diversity and organization of plankton communities, demonstrating that ecological functioning and taxonomic richness need not recover together."
    },
    {
      "heading": "Environmental mechanisms and extinction selectivity",
      "text": "Mass extinctions arise when environmental change exceeds organisms’ capacity to persist, migrate, or adapt. Different triggers can converge on similar stresses. Large-scale volcanism and asteroid impacts, for example, can both alter climate and primary productivity, although their immediate effects and characteristic timescales differ. Plate tectonics provides a slower background influence by changing continental positions, ocean gateways, and the distribution of shallow seas. Establishing a cause requires evidence that the proposed disturbance occurred at the right time and produced stresses consistent with the observed biological losses.\n\nTemperature changes influence physiology directly. Warming can increase metabolic oxygen demand while reducing the amount of oxygen dissolved in seawater. Animals may therefore encounter a shrinking zone in which both temperature and oxygen remain tolerable. Changes in ocean circulation and the decomposition of organic matter can intensify this mismatch. Cooling creates different pressures, including habitat shifts, ice expansion, and contraction of warm-water environments. The rate of change matters because migration and evolutionary adaptation require time and suitable destinations.\n\nOcean acidification results when added carbon dioxide changes seawater carbonate chemistry. Dissolved carbon dioxide shifts chemical equilibria, generally increasing hydrogen-ion concentration and reducing carbonate-ion availability. This can make skeleton construction more difficult for some organisms and promote dissolution under sufficiently undersaturated conditions. Responses vary with physiology, mineral composition, habitat, and the speed of the chemical disturbance. Acidification should not be inferred solely from a carbon-isotope excursion; stronger interpretations combine multiple chemical proxies with biological patterns.\n\nHabitat loss can amplify chemical and climatic stress. A sea-level fall may expose broad continental shelves, reducing living space for shallow-water communities. Sea-level rise can create new shallow habitats but also carry oxygen-poor water into previously habitable areas. On land, shifts in rainfall, fire regimes, and vegetation can fragment habitats and interrupt food supplies. These processes interact through feedbacks: loss of vegetation can increase erosion and nutrient runoff, potentially contributing to coastal oxygen depletion.\n\nSelectivity is studied by comparing victims and survivors. Geographic range is often important because widespread species are less likely to lose every suitable habitat simultaneously. Restricted endemism, narrow environmental tolerances, and specialized feeding can increase vulnerability, but their effects differ among events. A trait advantageous during one crisis may offer little protection during another. Large body size, for example, is not a universal predictor across all ancient extinctions.\n\nResearchers also distinguish the initiating trigger from the killing mechanisms. A volcanic province is a source of gases; warming, oxygen depletion, and acidification are pathways through which those emissions affect organisms. This separation allows competing explanations to be tested against independent evidence rather than merely associating an extinction with a nearby geological event."
    },
    {
      "heading": "Recovery and evolutionary consequences",
      "text": "Recovery has several meanings. The abundance of surviving organisms can increase rapidly, primary productivity can resume, and some ecological functions can return while species richness remains low. Conversely, species counts may rise before complex community structure is rebuilt. A return to the exact pre-extinction ecosystem is generally impossible because extinct lineages are permanently lost and environmental conditions continue to change. Post-extinction recovery is therefore a process of ecological reconstruction and evolutionary replacement, not simple restoration.\n\nImmediately after a crisis, communities may be dominated by a few widespread, disturbance-tolerant organisms. Such “disaster taxa” can become abundant because competitors, predators, or habitat-forming species have disappeared. Their prevalence does not necessarily indicate healthy or fully functioning ecosystems. Fossil assemblages may show low diversity, small body sizes, simplified food webs, or reduced disturbance of sediment by animals. These patterns vary with the nature of the crisis and the environments sampled.\n\nSurviving lineages can undergo adaptive radiation, diversifying into ecological roles that have become available. However, ecological opportunity alone does not determine the outcome. The surviving organisms must possess suitable traits, reach the relevant habitats, and persist through continuing environmental stress. Diversification can be delayed by persistent warming, oxygen depletion, loss of habitat structure, or disruption of interactions among species. Reefs are particularly demanding to rebuild because they depend on complex associations among framework builders, symbionts, grazers, and other organisms.\n\nEvolutionary consequences can extend for tens of millions of years. The end-Permian extinction contributed to a broad reorganization of marine communities, including changes in the relative importance of brachiopods, molluscs, and other groups. The end-Triassic event altered the composition of terrestrial vertebrate faunas before the Jurassic expansion of dinosaurs. After the end-Cretaceous extinction, mammals and surviving birds diversified into many ecological roles. In each case, groups that later became prominent had evolutionary histories predating the crisis.\n\nApparent recovery times depend on what is measured and the resolution of the record. A local community may recover sooner than global diversity, while immigration can restore regional richness without new species evolving. Some taxa disappear from sampled rocks and later reappear; these are called Lazarus taxa. Their absence may reflect survival in unsampled refuges or low abundance rather than true extinction followed by re-evolution. Distinguishing these possibilities requires careful attention to geography and preservation.\n\nMass extinctions also alter which evolutionary experiments remain available. A lineage eliminated during a brief crisis can no longer contribute descendants, even if it had been successful for millions of years. Consequently, long-term evolutionary history reflects both ordinary processes such as natural selection and rare disruptions that restructure the set of surviving lineages."
    },
    {
      "heading": "Present-day biodiversity loss and the sixth-extinction question",
      "text": "The phrase “sixth mass extinction” describes the possibility that human-driven biodiversity loss is producing, or could produce, a crisis comparable in magnitude to the Big Five. Modern pressures include conversion and fragmentation of habitats, direct exploitation, introduced species, pollution, and climate change. These drivers often interact. A population reduced by habitat loss, for example, may become more vulnerable to drought, introduced predators, or disease. Marine systems additionally face pressures such as fishing, warming, deoxygenation, and acidification.\n\nModern evidence differs fundamentally from the fossil record. Historical observations, surveys, museum specimens, and monitoring can document population changes at annual or decadal resolution. Fossil estimates generally integrate much longer intervals and disproportionately represent organisms with preservable hard parts. Comparisons must therefore account for differences in timescale, taxonomic coverage, and what counts as an observed extinction. A contemporary population decline is biologically important but is not itself the global extinction of a species.\n\nExtinction rates are sometimes expressed as extinctions per million species-years. One extinction per million species-years is equivalent to one species disappearing annually from a hypothetical pool of one million species. Estimates of long-term background rates commonly lie around 0.1–1 extinction per million species-years, although they vary among groups and methods. Many studies find that recent rates in well-studied groups exceed background estimates substantially. The precise multiplier depends on the chosen baseline, treatment of uncertain extinctions, and completeness of observations.\n\nThe 2019 global assessment of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services estimated that around one million animal and plant species were threatened with extinction. That figure refers to risk, not to one million confirmed extinctions. Likewise, a species classified as threatened under the IUCN Red List has not necessarily disappeared, and a species classified as extinct in the wild may persist in captivity. Maintaining these distinctions is essential when describing the scale of the crisis.\n\nDocumented modern losses have not yet established the disappearance of 75 percent of all species, the conventional threshold often used for a completed mass extinction. Nevertheless, elevated extinction rates, widespread population declines, and extensive habitat transformation provide strong evidence of a major ongoing biodiversity crisis. Whether the term “sixth mass extinction” describes a process already underway or a possible eventual outcome partly depends on the operational definition and timescale being used.\n\nPresent-day losses also include changes that species counts alone miss. The disappearance of populations reduces genetic diversity, geographic range, and ecological interactions before global extinction occurs. Long-lived individuals may persist after successful reproduction has ceased, creating an extinction debt: future losses resulting from earlier environmental damage. These lagged responses complicate any estimate based only on species that have already been confirmed extinct."
    }
  ],
  "links": [
    "extinction",
    "biodiversity",
    "species",
    "paleontology",
    "geology",
    "evolution",
    "speciation",
    "radiometric-dating",
    "isotope",
    "carbon-cycle",
    "ice-sheet",
    "forest",
    "carbon-dioxide",
    "oxygen",
    "photosynthesis",
    "plate-tectonics",
    "adaptation",
    "ocean-acidification",
    "feedback",
    "endemism",
    "natural-selection",
    "habitat",
    "climate-change",
    "iucn-red-list",
    "genetic-diversity"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}