Mass extinctions

Extinction is a normal part of evolution, but its pace is not constant. Most lineages disappear against a background of continuing origin and loss. A mass extinction is different: biodiversity falls unusually quickly on a global scale, across several environments and unrelated groups. The five most severe crises recognised in the well-sampled Phanerozoic fossil record are commonly called the Big Five.

The phrase is useful, but it can mislead. The events did not share one trigger, one duration or one biological result. The end-Ordovician crisis was closely tied to glaciation and sea-level change; the end-Permian and end-Triassic events were associated with immense volcanic provinces; the end-Cretaceous crisis preserves a global impact layer. Late Devonian losses occurred in several pulses rather than at one clean instant.

How scientists recognise a mass extinction

A fossil group vanishing from one quarry is not enough. Its local rock record may end because sediment stopped accumulating, erosion removed younger beds, the environment moved elsewhere or the organism became too rare to find. Researchers compare many sections, continents and habitats, then test whether last appearances cluster within a narrow interval.

Taxonomic counts are also corrected for sampling. A thick, accessible interval with many fossil-bearing formations will usually yield more species than a thin or poorly exposed one. Statistical methods estimate how much apparent decline comes from missing rock, uneven collecting or uncertain species names. The result is a range of plausible losses, not a perfectly known percentage.

The Signor–Lipps effect creates another difficulty. The last fossil of a species almost always lies below its true extinction point because fossilisation is rare. Sparse sampling can therefore smear an abrupt extinction downward through several layers. Exceptionally dense records, such as plankton across the K–Pg boundary, can narrow that gap.

Why a geological boundary is not always the extinction itself

Formal divisions on the geological time scale are anchored to an agreed point in a reference section or, for some Precambrian divisions, to a numerical age. A mass extinction may help identify a boundary, but the boundary is a convention for correlation. Biological losses can begin below it, peak close to it and continue above it.

The end of the Devonian illustrates the distinction. The Late Devonian crisis includes the Kellwasser events well before the formal Devonian–Carboniferous boundary and the later Hangenberg event close to that boundary. Calling all of them “the Devonian extinction” hides their spacing and potentially different mechanisms.

End-Ordovician: ice, falling seas and an oceanic reversal

Near the end of the Ordovician, most complex animal life occupied marine shelves. Expansion of an ice sheet over Gondwana cooled the climate and locked water on land. Sea level fell, shrinking the shallow habitats that contained much of marine diversity. Brachiopods, bryozoans, trilobites, graptolites and reef communities suffered strongly, but the pattern differed among regions and water depths.

The crisis had at least two major pulses. The first accompanied cooling and habitat loss. As the ice retreated, warming and rising seas did not simply reverse the damage: circulation and oxygen conditions changed, producing a second selective pressure. This sequence explains why “an ice age killed everything” is too simple.

Late Devonian: several crises, not one sudden boundary

Devonian seas supported extensive reef systems, armoured fishes, early sharks and diverse shelled invertebrates. On land, forests expanded and soils deepened. During the later Devonian, reef builders declined sharply and repeated low-oxygen episodes spread through marine basins. The Kellwasser events near the Frasnian–Famennian transition and the younger Hangenberg event are the best-known pulses.

Several mechanisms may have interacted. Weathering and nutrient runoff from increasingly complex terrestrial vegetation could have stimulated algal productivity and oxygen loss. Volcanism, climate swings, changes in sea level and ocean circulation also have supporting evidence in different records. No single proposed trigger yet explains every pulse equally well.

End-Permian: the deepest biological crisis

The end-Permian event, close to 251.9 Ma, caused the greatest known loss of marine biodiversity and a severe restructuring of terrestrial ecosystems. It marks the boundary between the Palaeozoic and Mesozoic eras, but extinction intensity and recovery were not identical in every environment.

The timing overlaps immense Siberian Traps eruptions. Magma and heated sedimentary rocks released carbon dioxide and other gases. Rapid greenhouse warming reduced oxygen solubility, accelerated continental weathering and altered nutrient delivery. Ocean acidification, deoxygenation and locally sulphidic water compounded the stress. On land, heat, aridity, vegetation change, erosion and unstable food webs affected animals and plants.

Recovery was prolonged and uneven. Some Early Triassic communities were species-poor, while repeated warming and oxygen stress interrupted rebuilding. Surviving lineages did not merely refill empty roles. Their descendants formed new communities, including the archosaur-dominated worlds described in the Triassic guide.

End-Triassic: volcanism during the breakup of Pangaea

Near 201.4 Ma, extinction removed the last conodonts, heavily reduced ammonites and reef communities, and eliminated several prominent terrestrial reptile lineages. Early dinosaurs survived, but the event should not be described as a planned “clearing of the way”. Survival was selective, and later dinosaur expansion followed within altered ecosystems.

The crisis coincided with eruptions of the Central Atlantic Magmatic Province as rifting opened the future Atlantic. Repeated volcanic pulses injected carbon into the atmosphere and are recorded by carbon-isotope disturbances. Rapid warming and ocean acidification provide a stronger combined explanation than lava flows alone, which covered only limited parts of the planet.

K–Pg: a sharp layer and a global aftermath

The Cretaceous–Palaeogene boundary at 66 Ma preserves the clearest physical fingerprint among the Big Five. A thin layer enriched in iridium, impact spherules and shocked minerals appears at sites around the world. Its age matches the Chicxulub crater beneath the Yucatán region.

The impact caused devastating regional effects, but the global killing mechanism came largely through the atmosphere. Dust, sulphate aerosols and soot reduced sunlight, cooled the surface and suppressed photosynthesis. Food webs collapsed on land and in the ocean. All non-avian dinosaurs disappeared, while birds survived; ammonites, mosasaurs and many plankton groups also vanished.

Deccan Traps volcanism affected climate during the same broad interval and may have stressed ecosystems. Yet the globally synchronous ejecta layer, crater and abrupt extinction signal identify Chicxulub as the principal cause of the terminal event. The evidence and survivor patterns are examined in detail in why dinosaurs went extinct. The Cretaceous guide places the boundary in its longer ecological setting, while the Palaeogene guide follows the recovery.

Why some organisms survived

Mass extinctions are selective, not random erasures. Geographic range, habitat, body size, food requirements, growth rate and reproductive strategy can all influence risk. The helpful combination changes with the crisis. A trait that buffers brief darkness may not protect against sustained ocean warming.

Survival also has a historical component. A lineage must already possess suitable traits and live in a tolerable place when the disturbance arrives. Evolution does not anticipate an asteroid or volcanic province.

Recovery is not a return to the old world

Species counts can begin rising before ecosystems regain their previous complexity. Early recovery communities may contain abundant opportunists but few large predators, reef builders or specialised interactions. Different regions recover at different rates, and evolutionary novelty can appear alongside long ecological instability.

Each crisis redirected the history of life. End-Ordovician survivors rebuilt marine communities; later Devonian losses ended the dominance of several reef systems; the end-Permian event preceded new Mesozoic ecosystems; the end-Triassic crisis was followed by dinosaur diversification; K–Pg survival set the stage for Cenozoic radiations of birds and mammals. None of these outcomes was inevitable.

Beyond the Big Five

The fossil record contains many other severe turnovers. The Capitanian crisis preceded the end-Permian event, several oceanic anoxic events disrupted Mesozoic seas, and large Pleistocene animals disappeared unevenly near the end of the Quaternary glacial cycles. Ranking depends on the groups counted, the time window and the completeness of the record.

Modern extinction rates and population declines are elevated by habitat destruction, overexploitation, invasive species, pollution and rapid climate change. “Sixth mass extinction” is a warning grounded in measured trends, not a claim that the present has already produced a finished geological boundary identical to the Big Five. The future magnitude depends strongly on actions taken now.