The last day of the dinosaursROEN

The Chicxulub asteroid, minute by minute

The last day of the dinosaurs

Sixty-six million years ago an asteroid as wide as a large city fell into the sea that covered what is now the Yucatán peninsula. By the usual estimate, about three quarters of all species died out. The globe is the Earth as it then was, and the clock counts time from the impact.

For example, London: the rock beneath it was 6,500 km from the impact. The first tremor arrived after 9 min 57 s, and the fragments thrown from the crater after about 24 min.

Scroll to move the clock

IThe day before

Land and sea at the end of the Cretaceous

The Atlantic was narrow, the sea covered part of today's land, and India was an island.

The thin lines are today's coastlines, moved to where their rock stood then. That is why the page speaks of the rock beneath a town: the towns, and often the dry land itself, came much later.

Half a day before

Europe was an archipelago

Most of Europe was islands in the sea. On land, Britain has no rock from that day: the youngest chalk dates from well before the impact, and in Northern Ireland about ten million years are missing between the chalk and the lavas above it.

Six hours before

Lava was flowing in India

India was being covered by the lavas of the Deccan, known as the Deccan Traps: more than a million cubic kilometres, before and after the impact. Two dating studies published in 2019 disagree on the tempo of the eruptions, and their part in the extinction is not settled.

An hour before

The place and the season

The asteroid fell into a sea that covered a platform of limestone and sulphur-rich rock, where the north of the Yucatán peninsula is today. It was spring in the northern hemisphere, as the bones of fish that died that day in North Dakota show.

IITen minutes before

An asteroid taller than Everest

Its size is not known directly: the studies use a body between 10 and 17 km in diameter. Even the smallest is taller than Everest.

A 2024 study of ruthenium, a rare metal preserved in the thin layer of clay the impact left all over the world, indicates a carbon-rich asteroid that formed beyond the orbit of Jupiter.

Six seconds before

In the simulations the asteroid travels at 12 or 20 km a second. If it strikes at 20 km a second, it covers the distance from the Moon in about six hours and crosses the atmosphere in six seconds. It came from the north-east on a steep path, 45–60° above the horizon.

IIIT + 1 s

The first second

The energy released is of the order of 1023 joules: between a few tens of millions and a hundred million megatons of TNT.

T + 25 s

A hole 100 km wide

In about half a minute the crust is pushed down and aside. What is left is a deep, bowl-shaped hole about 100 km wide, its walls covered with molten rock.

The drawing is to scale, with Everest beside it. Widths and times are taken from the studies; depths are approximate.

T + 1 min 40 s

A mountain that exists for a few minutes

The floor of the hole rises again. The shattered rock flows like a liquid and climbs, at the centre, kilometres above the surface. Between the third minute and the fifth, the mountain collapses towards the edges.

T + 10 min

The crater takes its final shape

What remains is a basin 180–200 km wide with a ring of mountains inside it, made of rock brought up from 8–10 km down. A borehole drilled into that ring in 2016 found 130 m of melted rock and rubble laid down on the first day.

IVT + 2 min 30 s

The whole Earth shakes

The seismic waves that cross the inside of the planet are faster than any wave at the surface. The thin white ring is where the first wave comes up from the deep; the yellowish one is the second, slower. The shock is estimated to have equalled an earthquake of magnitude 10–11 or more. The largest earthquake ever recorded, in Chile in 1960, had a magnitude of 9.5.

T + 9 min 57 s

The first tremor: London

The first wave reaches the rock beneath it after 9 min 57 s. The time is computed with the model of the Earth's interior that seismologists use to locate today's earthquakes.

T + 20 min

After 20 minutes the first wave has reached the opposite point of the globe. There, later, when the waves converge from every direction, one simulation indicates peak ground movements of about 4 m.

VT + 5 min

The rock thrown from the crater falls back

The clock goes back to minute five. The crater threw molten rock above the atmosphere at more than 5 km a second. The fragments travel round the planet on ballistic paths, like projectiles, and fall back hot. The bright points on the globe are those fragments; the red area is where they have begun to come down.

For each distance the globe shows the path of the slowest fragment that can reach it.

T + 15 min

The fish of North Dakota

At Tanis, about 3,000 km from the crater, fish were preserved with glass spherules in their gills: droplets of molten rock that cooled in flight. The researchers who described the site calculated that spherules arrived above it 13 to 25 minutes after the impact. The flight to Tanis computed on this page lasts 15 minutes.

T + 24 min

The fragments: London

The fragments begin to enter the atmosphere here after about 24 min. What fell then is still in the rock: a layer of clay a few millimetres thick, rich in iridium, a metal rare in the Earth's crust and more common in meteorites. The diamonds mark places where it has been measured. None is known on land in Britain or Ireland; the nearest in the table below are Stevns Klint in Denmark and Bidart in France.

T + 42 min

How hot the sky became

In this page's calculation, after 42 minutes the fragments are falling at the opposite point too. One study shows that as they fell they exposed the whole Earth, for several hours, to infrared radiation, that is to heat, enough to set forests alight. Another shows that the falling fragments shaded the ground and only a few minutes were dangerous. Charcoal was found in the crater; at six sites in North America almost none was found. How much of the forests burned has not been established.

VIT + 1 h

The wave leaves the gulf

In a simulation published in 2022, ten minutes after the impact and 220 km from the centre, the wave is 1.5 km high. In the open Gulf of Mexico it is generally more than 100 m high; after an hour it passes into the Atlantic. Its energy was up to 30,000 times that of the 2004 Indian Ocean tsunami.

The pale line on the globe is computed on this page: the shortest time in which the wave could reach each place, given the depth of the seas of that time.

T + 4 h

After four hours the wave passes between the two Americas, then separated by sea, and enters the Pacific. The rings of the seismic waves are long gone. The broad, pale white ring is the pressure wave in the air.

T + 5 h 49 min

The wave in the air

In 1883 the pressure wave from Krakatoa went round the Earth at about 310 m a second. At the same speed, the wave from the impact passes over the rock beneath London after about 5 h 49 min. How strong it still was at such distances is not known.

T + 48 h

Within two days the wave has reached most coasts. The Gulf of Mexico, the North Atlantic and the South Pacific were hit hardest. The Tethys, the sea that then covered southern Europe, was largely shielded.

VII

That day: London

  1. T + 0The impact, 6,500 km away. Nothing can be seen from here: the place is far below the horizon.
  2. after 9 min 57 sThe first tremor: the first seismic wave, arriving through the Earth's interior.
  3. after 18 minThe second seismic wave, slower.
  4. after 24 minFragments from the crater begin to enter the atmosphere overhead (computed for the slowest path).
  5. after 28 minThe waves that travel along the Earth's surface; at great distances they shake hardest.
  6. after 5 h 49 minThe pressure wave in the air, computed at the speed of the wave from the Krakatoa volcano in 1883.
  7. 1–2 yearsToo little light for photosynthesis, according to the climate models.

On the map of that time, the place is close to the shore.

The times are computed for the rock beneath the town, at the place where the plate model puts it. The page gives no arrival time of the sea wave for individual places.

VIIIT + 30 days

The sky goes dark

The impact lifted fine rock dust and, by one estimate, about 325 billion tonnes of sulphur into the upper atmosphere, and fires added soot. Researchers have not established which of them darkened the sky most. The models do agree that a year or two of dark and cold followed.

T + 1 year

Photosynthesis stops

In the dust model the light is too weak for photosynthesis for almost two years, and the average temperature falls by up to 15 °C. In the soot model there is no sunlight for more than a year. The one measurement in rock quoted here, made at a site in Texas, shows the sea cooled by up to 2 °C on average, with two drops of up to 7 °C.

T + 15 years

In the dust model the last particles settle after about 15 years. The light returns to a world with far fewer species.

IXT + 100 years

Who died out

By the usual estimate, about three quarters of all species died out. The counts by group are made in the places where rock from just before and just after the impact survives, which is why most hold for a single region.

T + 1,000 years

Who was left

In Montana nine in ten freshwater vertebrate species survived, and only one in eight of those on land: in fresh water, food chains start from dead matter; on land, from living plants. The lizards and snakes that survived were small. Today's birds descend, on one hypothesis, from a few groups that did not live in trees.

XT + 100,000 years

How long recovery took

Life was back inside the crater within years. Above the layer, in North America and New Zealand, the pollen of a rich flora is abruptly replaced by the spores of a few species of fern. On land, the recovery is recorded in exceptional detail in a sequence of rocks in Colorado.

T + 6 million years

The forests that grew in Colombia after the impact resembled today's tropical rainforests: flowering plants, with crowns that touch and shade the ground.

XIT + 33 million years

The continents keep moving

Half-way to today, the Atlantic is wider. The rock under London is moving with its continent and will reach the place where it is today.

Today

The crater lies under Yucatán

The crater lies half under the sea, half under land, covered by up to a kilometre of younger rock. At the surface the only trace is an arc of cenotes, natural water-filled sinkholes about 83 km from the centre.

Today

The dinosaurs that were left

There are more than 11,000 species of them, and they are called birds.

How long the waves and the fragments took to arrive: 8,771 towns and cities

The search on the page covers 8,771 towns and cities worldwide: the large cities and, for Britain, Ireland, Romania and Moldova, every town above 15,000 people. The table shows the largest in Britain and Ireland and a few elsewhere. Distance is measured over the globe, between the impact and the place where the rock under the town then stood.

Time from the impact to the arrival of the first seismic wave, the fragments (on the slowest path) and the pressure wave in the air; the last two rounded to the minute.
Town or cityDistance (km)First tremorFragmentsWave in the air
London6,5009 min 57 s24 min5 h 49 min
Birmingham6,4009 min 49 s24 min5 h 42 min
Dublin6,1009 min 30 s23 min5 h 26 min
Glasgow6,2009 min 40 s24 min5 h 34 min
Manchester6,3009 min 47 s24 min5 h 41 min
Sheffield6,4009 min 51 s24 min5 h 43 min
Leeds6,4009 min 50 s24 min5 h 43 min
Edinburgh6,3009 min 44 s24 min5 h 38 min
Liverpool6,3009 min 44 s24 min5 h 38 min
Bristol6,3009 min 46 s24 min5 h 40 min
Cardiff6,3009 min 43 s24 min5 h 37 min
Leicester6,4009 min 52 s24 min5 h 45 min
Bradford6,4009 min 49 s24 min5 h 42 min
Belfast6,1009 min 32 s23 min5 h 28 min
Coventry6,4009 min 51 s24 min5 h 43 min
Birkenhead6,3009 min 44 s24 min5 h 38 min
Nottingham6,4009 min 52 s24 min5 h 45 min
Reading6,4009 min 53 s24 min5 h 46 min
Kingston upon Hull6,5009 min 55 s24 min5 h 47 min
Preston6,3009 min 45 s24 min5 h 39 min
Swansea6,2009 min 40 s24 min5 h 35 min
Newcastle upon Tyne6,4009 min 50 s24 min5 h 43 min
Southend-on-Sea6,50010 min25 min5 h 52 min
Brighton6,5009 min 57 s24 min5 h 49 min
Derby6,4009 min 51 s24 min5 h 43 min

Where the layer can be seen today

The layer left by the impact has been found at hundreds of places. The 17 below appear on the globe as diamonds. The layer holds spherules (droplets of molten rock that cooled in flight), shocked quartz (crystals deformed by the pressure of the impact) and iridium, given in the table in parts per billion (ppb). Its thickness falls with distance: tens of metres within 500 km of the crater, from decimetres to ten metres at up to 1,000 km, 1–10 cm out to 5,000 km and millimetres beyond (Schulte et al., 2010).

The thickness bands above follow the distances estimated by Schulte, which differ from those computed here on the globe as it then was: larger for Europe, smaller for the Caribbean, the coast of the Gulf of Mexico and the sites in the southern hemisphere. The table shows both. An ODP site is a place where a core was drilled from the sea floor by the international Ocean Drilling Program.

PlaceDistance computed here (km)Distance estimated by Schulte (km)Thickness and contentsSource
Beloc, Haiti *900500decimetres to 10 m; spherules, shocked quartz; iridium 28 ppbSchulte et al. 2010, Table S1
El Mimbral, Mexico *1,000700decimetres to 10 m; spherules, shocked quartz; iridium 0.5 ppbSchulte et al. 2010, Table S1
ODP Site 1001, Caribbean Sea1,100600decimetres to 10 m; spherulesSchulte et al. 2010, Table S1
Brazos River, Texas *1,300900decimetres to 10 m; spherulesSchulte et al. 2010, Table S1
Starkville, Colorado *2,3002,2501–10 cm; spherules, shocked quartz, fern spores above; iridium 56 ppbSchulte et al. 2010, Table S1
Bass River, New Jersey2,4002,5001–10 cm; spherules, shocked quartzSchulte et al. 2010, Table S1
Tanis, North Dakota *3,0003,050a 1–2 cm iridium-rich layer (3.8 ppb) over a 1.3 m deposit laid down by water, which holds the spherulesDePalma et al. 2019
ODP Site 1259, off Suriname4,0004,5001–10 cm; spherules, shocked quartz; iridium 1.5 ppbSchulte et al. 2010, Table S1
Bidart, France *6,6009,500millimetres; iridium 6 ppbSchulte et al. 2010, Table S1
Caravaca, Spain *6,6008,200millimetres; spherules, shocked quartz; iridium 56 ppbSchulte et al. 2010, Table S1
Stevns Klint, Denmark *7,30010,200millimetres; spherules, shocked quartz; iridium 48 ppbSchulte et al. 2010, Table S1
El Kef, Tunisia7,7009,100millimetres; spherules; iridium 18 ppb. The international reference section for the boundary between the Cretaceous and the PaleogeneSchulte et al. 2010, Table S1
Gubbio, Italy7,9009,200millimetres; spherules, shocked quartz; iridium 8 ppb. Where the iridium was first reported, in 1980Schulte et al. 2010, Table S1
Byala, Bulgaria *8,8009,500millimetres; spherules, shocked quartz; iridium 6.1 ppbSchulte et al. 2010, Table S1
ODP Site 690, off Antarctica11,40011,000millimetres; iridium 1.5 ppbSchulte et al. 2010, Table S1
Woodside Creek, New Zealand *13,00010,500millimetres; spherules, shocked quartz, fern spores above; iridium 70 ppb, the most in the tableSchulte et al. 2010, Table S1
ODP Site 738, southern Indian Ocean15,10010,500millimetres; iridium 18 ppbSchulte et al. 2010, Table S1

* Position of the locality, not of the exact spot where the rock comes to the surface. For iridium the highest value measured is given, in parts per billion (ppb). In Schulte's table some rows group neighbouring places (ODP Sites 999 and 1001; 1258, 1259 and 1260; Stevns Klint and Nye Kløv; Gubbio and Petriccio; El Kef and Ellès).

What is computed and what is taken from studies

Where each town was

Positions 66 million years ago are computed with the PALEOMAP global plate model (Scotese, 2016), the same one that underlies the map of land and sea (Scotese and Wright, 2018; the map for 65 million years, in steps of one degree, about 110 km). The distance to the impact was checked for ten places on six continents against other reconstructions (Müller 2022 and Merdith 2021, which give the same distances at this age, and Seton 2012): they differ by 3.4% at most. Latitude differs more from one model to another, which is why the page does not show it.

For towns on volcanic islands younger than the impact (Iceland, Hawaii, the Canaries and others), the page says the island did not yet exist. “Under the sea” and “on land” are read from the maps for 65 and 70 million years, combined for the moment of the impact, at their step of one degree; “close to the shore” means the neighbouring squares of the map differ.

The shaking

The times of the first and second seismic waves are those of the waves that arrive first in the standard model of the Earth's interior used in seismology (ak135), for a source at the surface. Distances are measured in degrees over the globe: 90° is a quarter of the way round the Earth, about 10,000 km. The first wave is the compression wave, known as P. It arrives directly out to 98° from the crater. Between 98° and 159° a much weaker wave that bends round the core arrives first; beyond that, a wave that crosses the core, which is why the white ring pauses shortly before the opposite point. The second wave is the shear wave, known as S; beyond 83° it is an S wave that crosses the core. The waves that travel along the surface are taken at 3.8 km a second. A check: for Tanis, 3,000 km away, the calculation gives, rounded, 6, 10 and 13 minutes, the values published by DePalma and colleagues (2019).

The fragments

For each distance the page computes the ballistic flight with the lowest launch speed that reaches it, on a spherical Earth without air or rotation. Faster fragments on flatter paths could arrive sooner; most arrived later. For Tanis the result is 15 minutes; the researchers who described the site calculated 13 to 25 minutes for spherules. The arcs on the globe carry the computed times, but their shape is drawn in simplified form. How long their fall heated the sky, and how strongly, has not been established (Robertson et al., 2004; Goldin and Melosh, 2009).

The wave

The line on the globe is the quickest path of a tsunami, whose speed depends only on the depth of the water: √(g·h), where g is the acceleration of gravity and h the depth. The result, 27 hours for 95% of the sea surface and 43 at most, agrees with the milestones of the simulation by Range and colleagues (2022): one hour to leave the gulf, four to reach the Pacific, 48 to reach most coasts. Wave heights are taken from that study alone; the page computes no heights and no arrival time of the wave for individual places.

The wave in the air

The speed of 310 m a second is the one that best reproduces the barometer records of the wave from Krakatoa in 1883, collected in the Royal Society's report of 1888 (the calculation is on the page about that wave). For the impact it is an assumption: no measurement exists.

The crater in section

Widths and times follow the studies cited: a cavity 40–50 km in radius within tens of seconds, central uplift between seconds 20 and 100, collapse between 160 and 300, final shape in about ten minutes. The depth of the hole and the height of the central mountain do not appear in the text of those studies; in the drawing they are approximate.

Claims the page does not repeat

No study was found to support a few widespread statements: that no animal above 25 kg survived (it is a remark made at a 1976 workshop about the land vertebrates then known), that more than 90% of planktic foraminifera (microscopic sea organisms with shells) died out, and that an earthquake lasted weeks or months (a conference abstract).

Sources

  • Collins et al. 2020, “A steeply-inclined trajectory for the Chicxulub impact”, Nature Communications.
  • Fischer-Gödde et al. 2024, “Ruthenium isotopes show the Chicxulub impactor was a carbonaceous-type asteroid”, Science.
  • Morgan et al. 2022, “The Chicxulub impact and its environmental consequences”, Nature Reviews Earth & Environment.
  • Alvarez et al. 1980, “Extraterrestrial cause for the Cretaceous-Tertiary extinction”, Science.
  • Gulick et al. 2013, “Geophysical characterization of the Chicxulub impact crater”, Reviews of Geophysics.
  • Morgan et al. 2016, “The formation of peak rings in large impact craters”, Science.
  • Riller et al. 2018, “Rock fluidization during peak-ring formation of large impact structures”, Nature.
  • Gulick et al. 2019, “The first day of the Cenozoic”, PNAS.
  • Pope et al. 1996, “Surface expression of the Chicxulub crater”, Geology.
  • During et al. 2022, “The Mesozoic terminated in boreal spring”, Nature.
  • DePalma et al. 2019, “A seismically induced onshore surge deposit at the KPg boundary, North Dakota”, PNAS.
  • Meschede et al. 2011, “Antipodal focusing of seismic waves due to large meteorite impacts on Earth”, Geophysical Journal International.
  • Schulte et al. 2010, “The Chicxulub asteroid impact and mass extinction at the Cretaceous-Paleogene boundary”, Science.
  • Robertson et al. 2004, “Survival in the first hours of the Cenozoic”, GSA Bulletin.
  • Goldin & Melosh 2009, “Self-shielding of thermal radiation by Chicxulub impact ejecta: Firestorm or fizzle?”, Geology.
  • Belcher et al. 2003, “Fireball passes and nothing burns”, Geology.
  • Range et al. 2022, “The Chicxulub impact produced a powerful global tsunami”, AGU Advances.
  • Senel et al. 2023, “Chicxulub impact winter sustained by fine silicate dust”, Nature Geoscience.
  • Bardeen et al. 2017, “On transient climate change at the Cretaceous-Paleogene boundary due to atmospheric soot injections”, PNAS.
  • Artemieva & Morgan 2017, “Quantifying the release of climate-active gases by large meteorite impacts with a case study of Chicxulub”, Geophysical Research Letters.
  • Vellekoop et al. 2014, “Rapid short-term cooling following the Chicxulub impact at the Cretaceous-Paleogene boundary”, PNAS.
  • Longrich et al. 2016, “Severe extinction and rapid recovery of mammals across the Cretaceous-Palaeogene boundary”, Journal of Evolutionary Biology.
  • Longrich et al. 2012, “Mass extinction of lizards and snakes at the Cretaceous-Paleogene boundary”, PNAS.
  • Longrich et al. 2018, “Late Maastrichtian pterosaurs from North Africa and mass extinction of Pterosauria at the Cretaceous-Paleogene boundary”, PLOS Biology.
  • Landman et al. 2014, “Ammonite extinction and nautilid survival at the end of the Cretaceous”, Geology.
  • Sheehan & Fastovsky 1992, “Major extinctions of land-dwelling vertebrates at the Cretaceous-Tertiary boundary, eastern Montana”, Geology.
  • Wilf & Johnson 2004, “Land plant extinction at the end of the Cretaceous: a quantitative analysis of the North Dakota megafloral record”, Paleobiology.
  • Field et al. 2018, “Early evolution of modern birds structured by global forest collapse at the end-Cretaceous mass extinction”, Current Biology.
  • Lowery et al. 2018, “Rapid recovery of life at ground zero of the end-Cretaceous mass extinction”, Nature.
  • Vajda et al. 2001, “Indication of global deforestation at the Cretaceous-Tertiary boundary by New Zealand fern spike”, Science.
  • Renne et al. 2013, “Time scales of critical events around the Cretaceous-Paleogene boundary”, Science.
  • Lyson et al. 2019, “Exceptional continental record of biotic recovery after the Cretaceous-Paleogene mass extinction”, Science.
  • Wilson Mantilla et al. 2021, “Earliest Palaeocene purgatoriids and the initial radiation of stem primates”, Royal Society Open Science.
  • Carvalho et al. 2021, “Extinction at the end-Cretaceous and the origin of modern Neotropical rainforests”, Science.
  • Sprain et al. 2019, “The eruptive tempo of Deccan volcanism in relation to the Cretaceous-Paleogene boundary”, Science.
  • Schoene et al. 2019, “U-Pb constraints on pulsed eruption of the Deccan Traps across the end-Cretaceous mass extinction”, Science.
  • Hull et al. 2020, “On impact and volcanism across the Cretaceous-Paleogene boundary”, Science.
  • BGS Earthwise, “Cretaceous-Palaeogene (K-T) boundary, Northern Ireland”, British Geological Survey.
  • Scotese & Wright 2018, “PALEOMAP paleodigital elevation models (PaleoDEMS) for the Phanerozoic”, Zenodo.
  • Scotese 2016, “PALEOMAP PaleoAtlas for GPlates and the PaleoData Plotter Program”, EarthByte.
  • Kennett, Engdahl & Buland 1995, “Constraints on seismic velocities in the Earth from traveltimes (ak135)”, Geophysical Journal International.
  • Natural Earth, “1:50m coastline”.
  • GeoNames, “cities15000, CC BY 4.0”.
  • IOC World Bird List, v. 15.2.
  • Müller et al. 2022, “A tectonic-rules-based mantle reference frame since 1 billion years ago”, Solid Earth.
  • Merdith et al. 2021, “Extending full-plate tectonic models into deep time”, Earth-Science Reviews.
  • Seton et al. 2012, “Global continental and ocean basin reconstructions since 200 Ma”, Earth-Science Reviews.
  • GPlates Web Service.
  • The Krakatoa air wave: the speed of 310 m/s, fitted to the tables of the Royal Society report of 1888.
  • USGS, “M 9.5, 1960 Great Chilean Earthquake”.
  • International Commission on Stratigraphy, “GSSP for the Danian Stage, El Kef”.
  • BGS Earthwise, “Chalk Group lithostratigraphy: East Anglia, Beacon Hill Grey Chalk Member”, British Geological Survey.

Drag the globe to turn it.