How Did Dinosaurs Go Extinct?

Sixty-six million years ago, an asteroid roughly 10-15 kilometres across hit the Yucatán Peninsula at about 20 kilometres per second. The impact released energy equivalent to perhaps 100 million megatons of TNT — many billions of times the yield of every nuclear weapon ever built, combined. The site it left behind, the Chicxulub crater, is buried under the Gulf of Mexico and is about 180 kilometres wide.

This is not a theory. It is as well-evidenced as any fact in geology.

What followed was one of the worst mass extinctions in the history of complex animal life. Approximately 75% of all species on Earth went extinct. The non-avian dinosaurs — every lineage that hadn’t evolved into birds — were gone. So were the pterosaurs, the mosasaurs, the plesiosaurs, the ammonites, and most of the large marine reptiles. It took tens of millions of years for global biodiversity to rebuild.

The Evidence

The case for an asteroid impact was made in 1980 by physicist Luis Alvarez and his geologist son Walter Alvarez. They had noticed an anomaly in the geological record: at the boundary between Cretaceous and Palaeogene sediments — now called the K-Pg boundary, dated to exactly 66 million years ago — there was a thin layer of clay with iridium concentrations about 30 times higher than normal background levels. Iridium is rare in Earth’s crust but relatively common in asteroids.

The iridium layer exists globally. It has been found in Europe, North America, New Zealand, the ocean floor, Antarctica. It is one of the most consistently replicated findings in geology. The Alvarezes proposed a large asteroid impact as the only mechanism that could distribute iridium globally in a thin layer simultaneously deposited everywhere.

The crater was found a decade later. Drilling into the subsurface of the Yucatán Peninsula in the early 1990s confirmed the existence of a massive impact structure — the Chicxulub crater — with an age that matches the K-Pg boundary precisely. Shocked quartz (quartz crystals deformed by extreme pressure, a signature of hypervelocity impacts), glass spherules called tektites (formed from rock melted by the impact and ejected into the atmosphere), and a global soot layer consistent with continent-scale wildfires were all found at the same horizon.

The fossil record is equally clear. Below the K-Pg boundary, non-avian dinosaur fossils are abundant. Above it, they are absent. This stratigraphic signature is sharp — not a gradual tapering over millions of years, but a boundary.

What the Impact Actually Did

The immediate consequences of the impact were severe for organisms in and near North America. The Chicxulub site was a shallow sea overlying sulfur-rich rocks. The impact vaporised both the asteroid and a large volume of those rocks, injecting enormous quantities of sulfur compounds, water vapour, and particulate debris into the stratosphere.

The thermal pulse from ejecta re-entering the atmosphere would have been intense enough to ignite wildfires across large areas. The soot layer in the geological record is consistent with fires that burned on a continental scale. For animals without shelter, the initial hours and days were catastrophic.

The longer-term effect was the “impact winter.” Fine particles and sulfate aerosols in the stratosphere reflected incoming sunlight, causing a precipitous drop in surface temperatures and, more critically, shutting down photosynthesis. Models suggest photosynthesis was severely disrupted for months to years. Plants died. The herbivores that depended on them starved. The carnivores that depended on the herbivores followed.

The oceans acidified from dissolved sulfur compounds, collapsing marine food webs that depended on calcifying organisms — the kind that build shells from calcium carbonate. The loss of phytoplankton cut off the base of marine food chains simultaneously with the terrestrial collapse.

Why Dinosaurs Specifically?

The extinction was brutally size-selective on land. Animals above roughly 25 kilograms had very poor survival rates. This is not coincidental. Large animals need more food, have longer generation times (meaning they can’t adapt as quickly through reproduction), and have smaller populations. When the food supply collapses suddenly, animals with high caloric demands have no buffer.

The non-avian dinosaurs were also almost entirely large. By the Late Cretaceous, the smallest non-avian dinosaurs were still bird-sized at minimum. There was no lineage of truly small, ground-dwelling, insect-eating non-avian dinosaurs that could slip through the bottleneck.

Birds — which are avian dinosaurs — survived. The bird lineages that made it through appear to have been small, could likely eat seeds (which remain viable in soil long after plants die), and could fly to locate scarce food sources. Many of the bird lineages alive today, including shorebirds and some early ancestors of modern songbirds, trace back to groups that crossed the K-Pg boundary.

Small mammals also survived, for similar reasons: small body size, omnivorous diets, ability to burrow and find shelter, short generation times allowing rapid population recovery. Crocodilians survived, probably because their slow metabolisms allowed them to go for extended periods without food, and aquatic environments provided some buffering against the surface catastrophe. Turtles survived for similar reasons.

The pattern makes sense if you understand that this wasn’t an event that killed everything — it was an event that made survival dependent on specific attributes that large dinosaurs happened not to have.

The Deccan Traps Question

This is where the science gets more contested. At roughly the same time as the Chicxulub impact, large-scale volcanic eruptions were occurring in what is now western India — the Deccan Traps, one of the largest igneous provinces on Earth. These eruptions released significant amounts of greenhouse gases and sulfur compounds over hundreds of thousands of years, and there’s genuine debate about whether they were causing ecological stress in the Late Cretaceous that may have weakened dinosaur populations before the impact.

Some researchers have argued that the impact itself may have intensified Deccan volcanism through seismic energy transfer. This is controversial.

The current consensus is that the Chicxulub impact was the primary driver of the K-Pg extinction — the timing, speed, and global reach of the extinction all point to a single sudden trigger rather than a gradual decline. But whether Deccan volcanism was a contributing factor remains an active research question. The evidence from some fossil sites suggests dinosaur diversity was already declining in the final million years before the impact; other sites suggest healthy populations right up to the boundary. The disagreement is partly a question of where you’re looking and how you interpret incomplete fossil records.

The Aftermath

The world that emerged after the K-Pg extinction was fundamentally different from the one before it. For the previous 160 million years, large terrestrial ecosystems had been dominated by dinosaurs. Within about 10 million years of the extinction, mammals had diversified explosively into the ecological niches that dinosaurs had vacated: large herbivores, predators, burrowers, gliders, and eventually everything in between.

The success of mammals after the extinction wasn’t inevitable. Mammals had been small, mostly nocturnal animals for most of the Mesozoic — not because of anything inherently limiting about mammalian biology, but because dinosaurs occupied every large-body niche and kept them there through competitive exclusion. When the dinosaurs disappeared, that constraint was removed, and mammals radiated into body sizes and ecological roles that hadn’t been available to them for over 150 million years.

That radiation eventually produced primates, and eventually us. The K-Pg extinction is, in that sense, one of the most consequential single events in the history of life on Earth — not because of what it ended, but because of what it made possible.