Sixty-five million years ago the world was a very different place; the land was dominated by giant dinosaurs, pterosaurs filled the skies and the seas swarmed with giant marine reptiles and ammonites. Then overnight (in geological terms) everything changed. The non-avian dinosaurs disappeared, as did the pterosaurs, ammonites and all the marine reptiles except turtles and sea-snakes (crocodiles have since returned to the seas), and a number of other, less famous groups, such as the reef-building rudist bivalves. In geological sections (i.e. layered rocks that were laid down during this time) the end of the Cretaceous is marked by a thin layer containing unusually high levels of the rare element iridium, and often grains of 'shocked quartz' — sand grains which show signs of having been deformed by some abrupt pressure shock. This is followed by a 'Fern Spike' — a layer in sedimentary rocks in which fern spores are more abundant than flower pollen. Ferns are good at rapidly colonizing disturbed ground. This boundary is known as the KT or sometimes KPg boundary, with KT standing for Cretaceous, T for Tertiary and Pg for Palaeogene. K is used instead of C for the Cretaceous to avoid confusion with the Carboniferous Period (also Cretaceous starts with a K in German).
Scientists have (predictably) been arguing about what caused all this for a long time. In the last thirty years two main theories have emerged, polarizing the debate. The Impact Hypothesis proposes that some huge extra-terrestrial object crashed into the Earth, wiping out much of the world's fauna and flora overnight, where as the Deccan Traps Hypothesis proposes that a series of huge volcanic eruptions in what is now India poisoned the atmosphere over a period of a few thousand years, causing the extinction over a period of a few hundred thousand years, which is indistinguishable from a geological perspective.
This month two papers have appeared which re-examine this debate, from rather different perspectives.
The July edition of Biology Letters contained a paper by a team led by Tyler Lyson of the Department of Geology and Geophysics at Yale University and the Marmarth Research Foundation in North Dakota, which reports the discovery of the brow horn of a ceratopsian dinosaur 13 cm below the KT boundary, as defined by pollen changes, in the Hell Creek Formation of North Dakota. This is important as no dinosaur fossils have been found in the 3 m bellow the boundary at Hell Creek, one of the few sections where the KT transition is recorded clearly, which has been taken as evidence that the dinosaurs had become extinct before the KT boundary, and that their extinction could not, therefore, have been caused by an impact at the boundary.
An artists reconstruction of Zuniceratops, a ceratopsian dinosaur with prominent brow horns. By dinosaur artist Eivind Bovor.
There are two of flaws in this argument.
Firstly dinosaurs are big, rare, fossils. It takes exceptional circumstances for an animal as large as a dinosaur to be buried and fossilized. Most dinosaur species are only known from a single specimen. As such they are not good stratigraphic markers (unlike plankton or pollen); there could have been plenty of dinosaurs around at the time the top 3 m of Cretaceous sediments were laid down, but none of them happened to enter the fossil record.
Conversely a single horn is not evidence of the presence of dinosaurs either. The Hells Creek formation is a river system in which it would have been quite possible for bones (or horns) to be washed out from where they were originally buried and deposited again somewhere else; these are known as 'reworked' fossils. Indeed dinosaur teeth have been found at Hell Creek 1.3 m above the KT boundary, which are generally assumed to be reworked. Only a reasonably intact dinosaur skeleton should be interpreted as evidence for the presence of living dinosaurs.
This leaves a number of possible ways in which this fossil could be interpreted, eventually leaving us none the wiser.
1) Dinosaurs lived happily till the end of the Cretaceous, when they were wiped out by a huge bolide impact. This is the latest fossil we have found.
2) Dinosaurs declined in numbers throughout the late Cretaceous, with their numbers being depleted by a series of environmental catastrophes caused by volcanism in the Deccan Traps, then finally wiped out by a particularly serious volcanic event at the end of the period. This is the latest fossil we have found.
3) Dinosaurs declined in numbers throughout the late Cretaceous, with their numbers being depleted by a series of environmental catastrophes caused by volcanism in the Deccan Traps, then finally wiped out by a bolide impact at the end of the period. This is the latest fossil we have found.
4) Dinosaurs declined in numbers throughout the late Cretaceous, with their numbers being depleted by a series of environmental catastrophes caused by repeated bolide impacts due to a swarm of objects hitting the Earth, then finally wiped out by a particularly large bolide impact at the end of the period. This is the latest fossil we have found.
5) Dinosaurs lived happily till the end of the Cretaceous, when they were more-or-less wiped out by a huge bolide impact. This is not the latest fossil we have found, a few persisted into the Palaeogene.
6) Dinosaurs declined in numbers throughout the late Cretaceous, with their numbers being depleted by a series of environmental catastrophes caused by volcanism in the Deccan Traps, they finally died out some time in the Palaeogene. This is not the latest fossil we have found.
7) Dinosaurs declined in numbers throughout the late Cretaceous, with their numbers being depleted by a series of environmental catastrophes caused by repeated bolide impacts due to a swarm of objects hitting the Earth, some persisted into the Palaeogene. This is not the latest fossil we have found.
8) Dinosaurs died out before the end of the Cretaceous due to volcanic activity in the Deccan Traps. The fossil is reworked.
9) Dinosaurs died out before the end of the Cretaceous due to a bolide impact. The fossil is reworked.
10) Dinosaurs did not die out before the end of the Cretaceous, but the fossil is still reworked.
There are (obviously) other variants of this that could be speculated on, but this is enough to make the point; the presence of a single fossil is not enough to establish the presence of dinosaurs at a given point in time, nor to make assumptions about their decline.
The second paper appeared in the July 7 edition of the journal Nature, and was by Steve Cande and Dave Stegman of the Scripps Institution of Oceanography. Cande and Stegman do not speculate on the extinction event directly, but rather the cause of the Deccan Traps volcanicity.
It is generally accepted that the Deccan Traps flood basalts are associated with the collision of India and Eurasia, but exactly why this happened is not immediately clear. Africa and Australia are both at different stages of colliding with Eurasia, and show no sign of producing volcanic activity on such a huge scale, and North America and Europe have collided at least twice in the past without doing so.
The Deccan Traps are a huge area of basalt, covering much of what is now northwest India. It is theorized that the outpouring of these basalts would have been a catastrophic event with a profound impact on the Earth's climate and atmosphere.
Cande and Stegman examined the role of volcanic hot-spots in the movement of the tectonic plates. Hot spots are thought to occur where hot plumes from deep within the Earth's mantle, possibly as deep as the mantle/core boundary, reach the surface. Since these plumes originate deep below the lithosphere they are not effected by the movement of the tectonic plates, thus as a plate rolls over a hotspot it can create a string of volcanoes at the surface, as seen in the Hawaiian or Canary Islands.
Cande and Stegman theorized that where such mantle plumes reach the surface at a divergent boundary between two tectonic plates then they could act as an additional driving force, pushing the plates apart. They then carefully analyzed available data on the movements of the plates during the Late Cretaceous and Early Palaeogene, and made a surprising discovery. The origin of the Réunion Hotspot occurred on the boundary between the African and Indian Plates, about 70 million years ago. This coincides with first outpourings of the Deccan Traps Flood Basalts, and an increase in the momentum with which India moved northwards into Eurasia, leading to the creation of the Himalayas.
The plume remains active today (though much less so) and is responsible for a string of volcanoes and sea mounts across the Indian Ocean, the Laccadive Islands, the Maldives, the Chagos Archipelago, Mauritius, Rodrigues and Réunion Island all derive from the hotspot.
The progress of the Réunion Hotspot.
Under this model the origin of the Réunion Hotspot becomes the ultimate cause of the mass extinction event at the end of the Cretaceous, the Deccan Traps being a secondary cause. This has serious implications for us. It is theorized that a Hotspot lies beneath Iceland, fueling its volcanicity. Certainly Iceland is unusual in that it forms part of the Mid-Atlantic Ridge (a divergent plate boundary) that has formed a substantial landmass (divergent plate boundaries do not normally do this), and one third of all basalt produced at the Earth's surface during recorded history has been produced in Iceland.
In 1994 Lawrence Lawver of the Institute for Geophysics at the University of Texas and Dietmar Müller at the Department of Geology and Geophysics (now the School of Geosciences) at the University of Sydney published a paper in the journal Geology in which they argued that the Iceland Hotspot may be over 130 million years old, predating the separation of North America and Eurasia, and that its progress can be traced across Iceland, Greenland, Axel Heiberg and Ellis Islands. Certainly in this time it has not produced anything comparable to the Deccan Traps Flood Basalts, so we have no reason to suspect that it might start now (the Réunion Hotspot seems only to have done this at its origin).
However not all scientists agree there is a hotspot below Iceland at all. Gill Foulger of the Department of Earth Sciences at the Durham University, who disputes the existence of mantle plumes, has published a number of studies on Icelandic geology which suggest that a hotspot is not necessary to explain Icelandic volcanism.
There are other potential hotspots which could trouble us in the future. In particular the Yellowstone Caldera in Wyoming, USA, is thought by some scientists to lie on top of a developing hotspot, which would have potentially devastating consequences - even without the production of flood basalts.
A 3D model of the Yellowstone Mantle Plume.
More worryingly still some scientists theorize that a mantle plume may be rising beneath Southern Africa, which would eventually lead to a massive volcanic episode, which would, in conjunction with the East African Rift (a developing divergent boundary), lead to the break up of the continent, with devastating consequences for the Earth's climate and atmosphere, though fortunately this is not thought to be likely for the next few million years.