Horizon (1964) s41e04 Episode Script
What Really Killed the Dinosaurs?
Everybody knows what wiped out the dinosaurs.
Sixty five million years ago it came from outer space.
The world changed in an instant, it became an Old Testament version of hell.
Scientists claim the whole world burned then they say dust blocked out the sun.
They say the earth plunged into deep freeze for months or years.
They say it drove the dinosaurs to extinction.
But maybe they've got it all wrong.
The impact theory says in effect that a rock fell out of the sky and killed everything, except for the things that it didn't kill.
I don't think that's much of an explanation.
Tonight, Horizon reveals the new science which challenges our understanding of what really killed off the dinosaurs.
The hills of north east Mexico are full of evidence of a mass murder, sixty five million years ago.
It's a crime that for over a decade, scientists thought they had already solved.
But then, a geology professor from Princeton discovered something that wasn't supposed to be there.
When I first saw it I couldn't believe it, but then when it dawned on me what it really meant, I jumped up.
It was the most exciting Eureka moment, it changed the whole story.
What Gerta Keller had found were just tiny balls of rock, but the implications were profound.
Then I realised that if I'm right, everyone else is wrong: then science is wrong, about one of the biggest catastrophes in history.
Gerta Keller's claim was so shocking that it has sparked one of the bitterest scientific controversies of recent years.
So when I think of Gerta Keller's so called evidence, it makes me sort of mad because none of it holds up.
It's based on arguments which to my mind are barely scientific.
The arguments are so bitter because Gerta Keller has re-opened the case which most others considered closed.
In doing so she has sparked a scientific civil war.
And all because she questioned the accepted theory about what wiped out the dinosaurs.
Scientists began to investigate a layer of rock formed sixty five million years ago.
It's seen in mines and rock outcrops around the world.
Below this layer there are lots of dinosaur fossils above it there are none.
It's called the KT boundary.
This is the KT boundary and because it's such a thin, sharp line we know something dramatic must have happened here, some catastrophe and until recently we have almost no clue what happened whatsoever here, so it remained a total mystery.
Then in 1979 they discovered a clue in the KT boundary.
A high concentration of an element called Iridium.
Such quantities are extremely rare on earth and usually come from outer space.
So as soon as you find ten thousand times more Iridium at the very moment when the dinosaurs disappear, you know somewhere on earth a very big impact must have happened by an asteroid or a comet.
In fact there was so much Iridium, scientists realised the asteroid must have been a staggering ten kilometres in diameter.
If the theory was right, the impact would have created a fireball, equivalent to ten billion Hiroshima bombs.
The shock wave alone would have destroyed all life for hundreds of miles around.
All they had to do to prove the theory was find the evidence.
It was Jan Smit who first discovered the traces of the impact.
Hidden in the KT boundary layers he found these tiny balls of rock.
Careful analysis revealed the telltale signs of their origin.
These are called spherules, they're actually made of round rock globules so we know they're condensed from a very hot vapour cloud and it also tells us some of the mineralogy in there tells us that these spherules, these globules originated at very high temperatures.
That's exciting, we know something hot happened and hot is associated with an impact.
The spherules were evidence that the fireball had vaporised billions of tonnes of rock.
In outer space the vapour condensed into tiny droplets which fell back all over the earth as white hot spherules.
Then scientists discovered another clue in the KT boundary, high levels of soot.
From America to New Zealand there seemed to be evidence of massive burning at the time of the impact.
It looked as if the world's forests had spontaneously ignited as the rain of spherules heated the atmosphere by up to a thousand degrees centigrade.
If we're looking at six hundred, a thousand degrees then this would instantly have ignited all the plant matter across the world and it just would have been sent up in flames.
The impact was also thought to have created a deluge of vicious acid rain.
The fireball had released chemicals which turned the water deadly.
It was suggested that the acid rain had a PH so low that it was like battery acid and if you had something that low in PH it would literally burn everything on the land from plants to dinosaurs to everything else.
Then there was the final clue from the KT boundary a high concentration of fern spores.
Ferns flourish whenever all other plants have been killed off by some environmental devastation.
So the predominance of fern spores known as a fern spike suggested something had wiped out every plant on the planet.
Fern spikes were found all over the world such as in New Zealand and this I think became stronger and stronger evidence that there was something like global darkness caused by an impact.
So the theory grew up that vast amounts of dust created by the impact must have blocked out the sun.
This could have plunged the world into freezing darkness for months or years.
Any dinosaurs which had escaped burning either froze or starved to death.
But the mammals were small and could burrow to escape the heat and the cold.
When conditions recovered they would emerge to inherit the earth.
It was an elegant theory and there was only one thing missing - the impact should have left a crater, two hundred kilometres across.
To be certain the theory was right they had to find it.
Alan Hildebrand who was based in Calgary in western Canada was determined to be the one to discover the crater.
He knew the spherule layers should be thicker the nearer he got to the crater itself so he followed the trail of the spherules.
Here in western Canada we have a layer at the Cretaceous-Tertiary boundary, it's about one centimetre thick on average.
In Colorado and New Mexico it's two or two and a half centimetres thick.
Still further south in the Caribbean as in Haiti it's about a half metre thick so by nineteen eighty nine we knew that the source crater had to be somewhere between North and South America.
Then Alan Hildebrand had a stroke of luck.
He happened to be looking at an old oil company map, he was looking at the Yucatan Peninsula in Mexico, near a village called Chicxulub, then he noticed something odd.
This is when I spotted: a horseshoe shape feature: we can see it better on this map.
The anomaly is a hundred and eighty kilometres across which was in the range of sizes predicted for this crater and laying on the Yucatan Peninsula, between North and South America it's within the region where the ejecta thickness says the crater should lie.
So this suggested that this might indeed be the crater that killed the dinosaurs.
Now they knew where the crater lay, the theory seemed to be proven.
But it wasn't quite so simple.
At Chicxulub the crater was buried a kilometre underground.
On the surface there was nothing to see.
To investigate the impact further would mean looking elsewhere.
Jan Smit rushed to Mexico.
He scoured the country for rock outcrops which were the same age as Chicxulub.
He hoped they would reveal the remaining secrets of the impact.
Finally he came to explore an area one thousand kilometres from the crater itself.
We came here as a sort of last gasp effort and finally the last rays of the sunset we found it.
It was a massive rock outcrop.
It seemed to be exactly the same age as Chicxulub.
At last the whole history of the impact could be read in a rock.
There were spherules from the crater and Iridium from the asteroid, but there was also something else.
What you see here, here is the spherule layer and if we go way up there you see the layers which we know are rich in Iridium and normally those two are together but now we see here, in between, huge masses of sand sandstone and the question was how did it get here? Sand is normally found only on the coast, but sixty five million years ago this whole region was under deep ocean.
The sandstone should not have been here.
According to Smit, it could only be evidence of another disaster caused by the asteroid.
And there must have been a lot of energy in the ocean here to bring all this sand down here and to me it was very obvious, very clear that this must have been a huge Tsunami wave.
Smit believed it was a Tsunami tidal wave like no other.
It would have travelled at hundreds of miles an hour It could have been a thousand feet high.
According to Smit, the Tsunami must have churned up billions of tonnes of sand from the coast and dumped it on the ocean bed.
This became the sandstone and when the sea level fell, the rock was exposed.
If Smit was right the entire outcrop was created in no time at all.
After the impact it took a few minutes for the molten material from the crater to get on the spot and deposit on the sea floor.
Then after a few hours we get huge Tsunami waves crashing in and they bring a lot of sand here to this place here.
Keller was fascinated by the Tsunami idea and followed Smit to Mexico.
Keller and her colleague, Wolfgang Stinnesbeck, investigated fifty rock outcrops in the same region.
The same sandstone feature Smit described occurred everywhere they looked, but somehow Smit's Tsunami analysis just didn't seem to fit.
It's a nice story but when you take a second look the story just doesn't stack up, there is something fishy about it, it just doesn't make sense.
In science they say there is nothing sadder than a beautiful theory ruined by a single awkward fact.
And Keller and Stinnesbeck thought they could see lots of awkward facts.
Even the first time that we came to this place we noticed there was something wrong with Jan Smit's story, just too many details that you have to explain some other way.
Their first awkward fact was evidence of ancient life, fossilised in the sandstone.
of a probably worm-like animal which well, ate the sediment and we have another layer here with dwelling structures of a crab.
It looked to them as if the lower layer of sandstone had been thoroughly colonised by worms over months or even years.
Only afterwards was the upper layer deposited and colonised by the crabs.
It seemed to them that there were two layers created at different times and not within a couple of days as Smit had suggested.
So this layer up here has nothing to do with the lower level, the two are separated by time, and by a considerable amount of time.
The next awkward fact was the existence of layers of muddy sedimentary rock, interspersed in the sandstone.
These they thought would have taken a long time to form.
They are made of fine grains which tend to build up slowly on the sea floor.
So these fine grains needed some time to settle, in fact an awful lot of time and in that case these sandstones would be separated by some time.
Then, near the bottom of the outcrop they discovered their third awkward fact - a layer of limestone.
Limestone is given away by its violent reaction to acid.
It can also take a very long time to form.
Finally, Keller and Stinnesbeck discovered their most awkward fact of all.
It lay eight metres below the outcrop.
It was another layer of spherules.
Chemical analysis suggested, these remains of tiny meteorites like the layer higher up could only have come from the Chicxulub crater.
For Keller and Stinnesbeck, this changed everything.
It was a Eureka moment, absolutely.
We danced, we jumped up and down and said we've got it.
They were excited because these spherules were eight metres below where any had been found before.
In geology, lower normally means further back in time, so for them it suggested an entirely different explanation for the whole rock formation and the extinction of the dinosaurs.
It was all to do with time.
After the Chicxulub impact, the spherules settled on the sea floor within minutes.
These were the spherules just discovered by Doctor Keller.
On top of the spherules there seemed to have been a period of normal sedimentation, eight metres of mud had built up.
According to Keller and Stinnesbeck this would have taken two hundred and seventy thousand years.
Next came the limestone layer, this too could have taken thousands of years to form.
Only then came the sand, washed in to form the sandstone.
Then those fine grained layers built up by sedimentation, that could have taken centuries.
Then worms apparently had months, maybe years in which to colonise some of the layers.
And only then, at the very top came the Iridium layer, which coincided with the extinction of the dinosaurs.
But by now Keller and Stinnesbeck calculated that three hundred thousand years had passed since the Chicxulub impact.
The Iridium must have come from an asteroid but it must have been a second impact, three hundred thousand years later for which we simply don't know where the crater is and it is that impact that probably finished off the dinosaurs.
If Keller was right, there hadn't been one impact but two.
The first, Chicxulub, would have been devastating enough It would have killed many dinosaurs but it did not drive them to extinction and it didn't create the Iridium layer.
According to Keller and Stinnesbeck, even as earth absorbed the Chicxulub impact, deep in outer space a second asteroid was hurtling toward the planet and this was the asteroid that left the telltale Iridium and finished off the dinosaurs.
At the end of 2001 an international team of scientists gathered near Chicxulub.
They used an oil rig to drill down over a kilometre into the heart of the crater.
They extracted hundreds of metres of rock cores, most of the rock had been pulverised by the Chicxulub impact but one section was undamaged.
It lay above the broken rock so it must have been created after the impact and just underneath the Iridium layer, so the question was how long had this section taken to form? If the mysterious rock had been formed slowly it would support Keller and Wolfgang Stinnesbeck.
What we can see here is the impact sediment which is all smashed off the rocks, we see bigger pieces and smaller pieces here, but then we find something completely different.
See this grey rock, this is a limestone and when I looked at the limestone with my head lens I noticed that the grains which are forming the limestone are very fine, so that would show to me that this sediment has been deposited under rather quiet conditions.
In other words, Stinnesbeck thought the rock had been formed slowly, by sedimentation.
There must have been a long time between the impact below the core and the arrival of the Iridium at the top.
Then they noticed something else in the core.
Traces of what looked like a green clay mineral.
Samples were sent for analysis by a technique called x-ray diffraction.
Every mineral has its own unique signature when exposed to x-rays.
The analysis revealed a clay mineral called Glauconite and Glauconite has one vital characteristic.
Glauconite needs many thousands of years to form, so it's indicating interruptions in the sedimentation of this limestone so it extends even more the time, the period of time which is needed for the deposition of this limestone.
Gerta Keller also had samples from the core.
She was looking for something quite different.
I just scanned them very rapidly and at first I didn't see anything, and then I began to see some shapes that looked familiar.
What Keller thought she had seen were microscopic plankton fossils called forams, but they were poorly preserved and it took an expert eye to make them out.
This is a 3D image of a perfect foraminifera.
This is what we found in the core, it is a cross section of the same species.
This is another 3D image, you see five chambers arranged in a circle, you can see the same chambers in the cross section.
Here is another species, each of the segments come to a pinched end and the same you can see on the cross section.
All these species went extinct at the same time as the dinosaurs, yet Keller had found their fossils in rock formed after the Chicxulub impact.
For Keller the meaning of the discovery was clear.
The fact that we find these species in the rocks above the Chicxulub impact rocks means that Chicxulub impact must predate the mass extinction by about three hundred thousand years and that just confirms what we already found.
Keller and Stinnesbeck thought they had won.
For them the evidence proved beyond doubt that Chicxulub was far too old to have caused the extinction of the dinosaurs.
But that was just one way to look at it.
When Jan Smit analysed the core, he saw something quite different.
He found evidence that the rock had been created quickly.
Look here for instance, you see here ripple marks, there's another layer of ripple marks, there are lines in the rocks, they all bear evidence for strong currents which implies Tsunami events.
Smit also examined the layers of green clay mineral.
He did his own x-ray diffraction analysis and found not Glauconite but a different mineral called Smectite.
Smectite can form quickly.
So they made another mistake, they misidentified the clay mineral with very severe implications.
And as for Gerta Keller's forams, Smit was in no doubt she had made another big mistake.
He took photographs through a microscope too - he found not a single fossil, just these inorganic crystals.
On these fuzzy pictures you can easily claim they are forams but I claim it's a fortuitous combination of those crystals in a very fuzzy picture.
If you take sharper pictures like these with a different method, you see they are totally different.
According to Smit, all the evidence suggested the core had been formed in just a few days by a Tsunami and that meant Chicxulub must be the killer of the dinosaurs.
This is still one and the same impact, we have the Chicxulub impact here, very quick sedimentation by Tsunami waves and at the very end we get the Iridium from the Chicxulub impact so the dinosaurs got extinct right here, as a consequence of the Chicxulub impact and not by a second impact on some unknown place in the world.
Jan Smit and Gerta Keller had fought each other to deadlock.
Smit still represents the view of the great majority of scientists in this field but there is now some support for Gerta Keller's position.
Neither side is about to back down.
We looked at all evidence and I refute everything.
There's not a single argument which holds ground.
He built his career on this impact Tsunami hypothesis and right now he's feeling this hypothesis crumble in front of his eyes.
But while two groups of scientists were obsessed with the debate about which was the guilty impact, another thought the whole thing was completely beside the point.
The objection was based on one simple observation earth has been hit many times and these impacts, even the really big ones, seldom coincide with mass extinctions.
For the very large craters that we think might have produced global effects, in some cases they don't line up with any extinction events whatsoever and in other cases extinction events, very well dated extinction events have occurred that are not associated with any impacts.
Because of this, some scientists had long doubted that any impact could have wiped out the dinosaurs and now there is new evidence which seems to reinforce those suspicions.
The new evidence relates to the forest fires that were said to have raged all over the world.
They seem to have been supported by soot in the KT boundary layers, but the problem with soot is that it can be blown thousands of miles from where the burning actually happened, so one researcher decided to investigate a more reliable kind of evidence charcoal! Large pieces of charcoal can't be transported great distances, they can't be picked up and lofted in smoke plumes or carried around the world so therefore if we can find charcoal at each KT boundary location we can be sure that burning occurred at each site.
In 2003, Claire Belcher scoured North America looking for charcoal.
She collected samples from eight KT boundary sites, from New Mexico in the south right up into western Canada.
In the lab she searched the samples for traces of charcoal, which show up white under the microscope.
She was amazed at what she discovered there was almost none.
What we did find however was a lot of this non charred plant material.
This was really surprising because it means that to have this amount of non charred plant material in the KT boundary rocks, there's no way that there can have been a globally extensive wild fire because it just simply wouldn't have left so much non charred material on the rock record.
Claire Belcher's results were consistent from Canada, all the way down to the southern United States, just two thousand kilometres from the crater.
Even here there was no evidence of burning.
If there was no burning across north America then how on earth could there be burning across the rest of the world because north America is so close to the impact site so basically it seemed there were no fires.
So the dinosaurs had not roasted.
A major predicted effect of the impact simply did not hold up.
And that raised a whole series of other questions.
What about the idea of acid rain, meant to have been as strong as battery acid, that should have destroyed the plants that many dinosaurs lived on and wiped out many sensitive animal species.
Biologist Dave Archibald has made a statistical study of the kinds of animals which went extinct and those that survived.
He found that many which should have died out because of acid rain but didn't.
This little tree frog like all amphibians is very susceptible to acid rain.
They have problems with their metabolism and finally they can't breed in the water.
If acid rain had occurred at the end of the Cretaceous, all amphibians should have disappeared.
They don't, almost all of them survived, as do all the turtles, the crocodiles, their cousins the alligators and most fishes.
This says that there can not have been battery acid strength acid rain at the end of the Cretaceous.
If there was no significant acid rain, it seemed another killing mechanism had been disproved.
And what about the so called impact winter? Dust was meant to have blocked out the sun that could have plunged the world into deep freeze for months or years, but again those little frogs got in the way.
Amphibians like all cold blooded creatures have problems dealing with sudden drops in temperature, especially if it lasts for many months or years.
If this had happened at the end of the Cretaceous, then many cold blooded organisms should have become extinct .
.
but they did not.
This tells us that there could not have been a long term drop in temperature.
So the dinosaurs did not freeze to death nor did they roast and nor did they die from the effects of acid rain.
There seemed to be so many holes in the impact theory, scientists began to dust down a more classical, evolutionary idea.
Maybe the dinosaurs had died out gradually and for completely different reasons.
Many scientists believe that about ten million years before the KT boundary, the dinosaurs' environment began to get markedly worse.
The lush coastal planes on which they depended began to turn arid, as sea levels fell dramatically.
Their natural habitat was disappearing.
They no longer have as much area to live in, they don't have as much food to eat, it's harder to get from one area to another.
Organisms are adapted for a certain type of environment and when those conditions disappear those organisms disappear.
Between ten million and three million years before the KT boundary, the dinosaurs seem to have suffered a dramatic decline.
One interpretation of the fossil record is that forty percent of dinosaur species went extinct.
And so we come up to about one to three million years before the end of the Cretaceous and we've got a group that seems to be vulnerable to any kind of extinction events that might be occurring at that time.
So it seems that by three million years before the impact, the dinosaurs were already in trouble.
The big question is what happened next? Did they stabilise or carry on dwindling? What is known is that their environment continued to worsen.
For about half a million years before the KT boundary, the world suffered one of its most destructive periods of volcanism ever.
Vast areas of land were ridden with eruptions.
A million cubic kilometres of lava spewed out in what are known as the Deccan Traps.
We're talking about the entire western part of India being a centre of volcanic activity with volcanoes and lava coming out, over a large area, over thousands of kilometres.
Volcanic greenhouse gases contributed to an apparently massive global warming.
Some data point to an average rise in temperature of eight degrees centigrade in the last half million years before the impact.
This would have been devastating.
We're talking about catastrophic effects in terms of changes in habitat, changes in rain fall patterns, changes in climate, all of these you can think of all of the things that are going on in the modern world magnified many many times, many many orders of magnitude indeed.
It is not clear what effect the Deccan Traps had on the dinosaurs.
The fossil record for this crucial period is difficult to interpret.
But the fossil record for other animals such as the ammonites is far better and using this some argue it's possible to infer what was happening to the dinosaurs.
Six million years prior to the KT boundary there were about twenty species of ammonites in the world's oceans.
Three million years before the KT boundary there were only fifteen or so and one million years prior to the KT boundary we have less than half of what we started out with, we have less than ten species so the extinction event has already been going on for millions of years.
The amazing thing is that we see the same pattern in the fish record, we see the same pattern in the terrestrial reptile record, we even see the same pattern in the mammal record.
All of these groups were undergoing an extinction event for millions of years and it would be absolutely amazing to me if dinosaurs weren't undergoing the same sort of extinction and indeed I think they were undergoing the same sort of long term extinction.
It seemed there was no role left for the asteroid impact or impacts at all.
The dinosaurs were doomed anyway: except there was still evidence of an environmental catastrophe at the KT boundary.
The awkward unavoidable fact that remains to support the impact theory was the fern spike.
The predominance of fern spores in the KT boundary layers suggests that all vegetation was destroyed for a time.
Those who are convinced that Chicxulub killed off the dinosaurs argue that the fern spike, caused by the impact, is still the key to understanding the mass extinction.
It's reasonable to think that all the standing vegetation all over the earth was destroyed, so we don't understand exactly how yet but certainly its evidence of a dramatic catastrophe in the terrestrial eco system from the impact.
Because the vegetation was not destroyed by wild fires, the most likely cause seems to be global darkness brought on by the impact dust and that would have been catastrophic for the dinosaurs.
Naturally if you're a plant eating dinosaur, all the plants that you're eating are dying so this of course is the end of you, and if you're a meat eating dinosaur preying on plant eating dinosaurs once they're gone, you're gone too.
The end of the dinosaurs now seems much more complicated than it once did.
It appears they were already dwindling under pressure from a worsening environment.
Massive volcanism and falling sea levels might have forced them to extinction, even if the asteroid had missed.
But there was a massive asteroid impact, either at Chicxulub or was it somewhere else entirely? And it does seem likely that this, at the very least finished off the remaining dinosaurs.
So it seems they were the victims of a lethal combination of circumstances.
Clearly dinosaurs were incredibly unlucky at the end of the Cretaceous to have all three things happen at the same time.
Sixty five million years ago it came from outer space.
The world changed in an instant, it became an Old Testament version of hell.
Scientists claim the whole world burned then they say dust blocked out the sun.
They say the earth plunged into deep freeze for months or years.
They say it drove the dinosaurs to extinction.
But maybe they've got it all wrong.
The impact theory says in effect that a rock fell out of the sky and killed everything, except for the things that it didn't kill.
I don't think that's much of an explanation.
Tonight, Horizon reveals the new science which challenges our understanding of what really killed off the dinosaurs.
The hills of north east Mexico are full of evidence of a mass murder, sixty five million years ago.
It's a crime that for over a decade, scientists thought they had already solved.
But then, a geology professor from Princeton discovered something that wasn't supposed to be there.
When I first saw it I couldn't believe it, but then when it dawned on me what it really meant, I jumped up.
It was the most exciting Eureka moment, it changed the whole story.
What Gerta Keller had found were just tiny balls of rock, but the implications were profound.
Then I realised that if I'm right, everyone else is wrong: then science is wrong, about one of the biggest catastrophes in history.
Gerta Keller's claim was so shocking that it has sparked one of the bitterest scientific controversies of recent years.
So when I think of Gerta Keller's so called evidence, it makes me sort of mad because none of it holds up.
It's based on arguments which to my mind are barely scientific.
The arguments are so bitter because Gerta Keller has re-opened the case which most others considered closed.
In doing so she has sparked a scientific civil war.
And all because she questioned the accepted theory about what wiped out the dinosaurs.
Scientists began to investigate a layer of rock formed sixty five million years ago.
It's seen in mines and rock outcrops around the world.
Below this layer there are lots of dinosaur fossils above it there are none.
It's called the KT boundary.
This is the KT boundary and because it's such a thin, sharp line we know something dramatic must have happened here, some catastrophe and until recently we have almost no clue what happened whatsoever here, so it remained a total mystery.
Then in 1979 they discovered a clue in the KT boundary.
A high concentration of an element called Iridium.
Such quantities are extremely rare on earth and usually come from outer space.
So as soon as you find ten thousand times more Iridium at the very moment when the dinosaurs disappear, you know somewhere on earth a very big impact must have happened by an asteroid or a comet.
In fact there was so much Iridium, scientists realised the asteroid must have been a staggering ten kilometres in diameter.
If the theory was right, the impact would have created a fireball, equivalent to ten billion Hiroshima bombs.
The shock wave alone would have destroyed all life for hundreds of miles around.
All they had to do to prove the theory was find the evidence.
It was Jan Smit who first discovered the traces of the impact.
Hidden in the KT boundary layers he found these tiny balls of rock.
Careful analysis revealed the telltale signs of their origin.
These are called spherules, they're actually made of round rock globules so we know they're condensed from a very hot vapour cloud and it also tells us some of the mineralogy in there tells us that these spherules, these globules originated at very high temperatures.
That's exciting, we know something hot happened and hot is associated with an impact.
The spherules were evidence that the fireball had vaporised billions of tonnes of rock.
In outer space the vapour condensed into tiny droplets which fell back all over the earth as white hot spherules.
Then scientists discovered another clue in the KT boundary, high levels of soot.
From America to New Zealand there seemed to be evidence of massive burning at the time of the impact.
It looked as if the world's forests had spontaneously ignited as the rain of spherules heated the atmosphere by up to a thousand degrees centigrade.
If we're looking at six hundred, a thousand degrees then this would instantly have ignited all the plant matter across the world and it just would have been sent up in flames.
The impact was also thought to have created a deluge of vicious acid rain.
The fireball had released chemicals which turned the water deadly.
It was suggested that the acid rain had a PH so low that it was like battery acid and if you had something that low in PH it would literally burn everything on the land from plants to dinosaurs to everything else.
Then there was the final clue from the KT boundary a high concentration of fern spores.
Ferns flourish whenever all other plants have been killed off by some environmental devastation.
So the predominance of fern spores known as a fern spike suggested something had wiped out every plant on the planet.
Fern spikes were found all over the world such as in New Zealand and this I think became stronger and stronger evidence that there was something like global darkness caused by an impact.
So the theory grew up that vast amounts of dust created by the impact must have blocked out the sun.
This could have plunged the world into freezing darkness for months or years.
Any dinosaurs which had escaped burning either froze or starved to death.
But the mammals were small and could burrow to escape the heat and the cold.
When conditions recovered they would emerge to inherit the earth.
It was an elegant theory and there was only one thing missing - the impact should have left a crater, two hundred kilometres across.
To be certain the theory was right they had to find it.
Alan Hildebrand who was based in Calgary in western Canada was determined to be the one to discover the crater.
He knew the spherule layers should be thicker the nearer he got to the crater itself so he followed the trail of the spherules.
Here in western Canada we have a layer at the Cretaceous-Tertiary boundary, it's about one centimetre thick on average.
In Colorado and New Mexico it's two or two and a half centimetres thick.
Still further south in the Caribbean as in Haiti it's about a half metre thick so by nineteen eighty nine we knew that the source crater had to be somewhere between North and South America.
Then Alan Hildebrand had a stroke of luck.
He happened to be looking at an old oil company map, he was looking at the Yucatan Peninsula in Mexico, near a village called Chicxulub, then he noticed something odd.
This is when I spotted: a horseshoe shape feature: we can see it better on this map.
The anomaly is a hundred and eighty kilometres across which was in the range of sizes predicted for this crater and laying on the Yucatan Peninsula, between North and South America it's within the region where the ejecta thickness says the crater should lie.
So this suggested that this might indeed be the crater that killed the dinosaurs.
Now they knew where the crater lay, the theory seemed to be proven.
But it wasn't quite so simple.
At Chicxulub the crater was buried a kilometre underground.
On the surface there was nothing to see.
To investigate the impact further would mean looking elsewhere.
Jan Smit rushed to Mexico.
He scoured the country for rock outcrops which were the same age as Chicxulub.
He hoped they would reveal the remaining secrets of the impact.
Finally he came to explore an area one thousand kilometres from the crater itself.
We came here as a sort of last gasp effort and finally the last rays of the sunset we found it.
It was a massive rock outcrop.
It seemed to be exactly the same age as Chicxulub.
At last the whole history of the impact could be read in a rock.
There were spherules from the crater and Iridium from the asteroid, but there was also something else.
What you see here, here is the spherule layer and if we go way up there you see the layers which we know are rich in Iridium and normally those two are together but now we see here, in between, huge masses of sand sandstone and the question was how did it get here? Sand is normally found only on the coast, but sixty five million years ago this whole region was under deep ocean.
The sandstone should not have been here.
According to Smit, it could only be evidence of another disaster caused by the asteroid.
And there must have been a lot of energy in the ocean here to bring all this sand down here and to me it was very obvious, very clear that this must have been a huge Tsunami wave.
Smit believed it was a Tsunami tidal wave like no other.
It would have travelled at hundreds of miles an hour It could have been a thousand feet high.
According to Smit, the Tsunami must have churned up billions of tonnes of sand from the coast and dumped it on the ocean bed.
This became the sandstone and when the sea level fell, the rock was exposed.
If Smit was right the entire outcrop was created in no time at all.
After the impact it took a few minutes for the molten material from the crater to get on the spot and deposit on the sea floor.
Then after a few hours we get huge Tsunami waves crashing in and they bring a lot of sand here to this place here.
Keller was fascinated by the Tsunami idea and followed Smit to Mexico.
Keller and her colleague, Wolfgang Stinnesbeck, investigated fifty rock outcrops in the same region.
The same sandstone feature Smit described occurred everywhere they looked, but somehow Smit's Tsunami analysis just didn't seem to fit.
It's a nice story but when you take a second look the story just doesn't stack up, there is something fishy about it, it just doesn't make sense.
In science they say there is nothing sadder than a beautiful theory ruined by a single awkward fact.
And Keller and Stinnesbeck thought they could see lots of awkward facts.
Even the first time that we came to this place we noticed there was something wrong with Jan Smit's story, just too many details that you have to explain some other way.
Their first awkward fact was evidence of ancient life, fossilised in the sandstone.
of a probably worm-like animal which well, ate the sediment and we have another layer here with dwelling structures of a crab.
It looked to them as if the lower layer of sandstone had been thoroughly colonised by worms over months or even years.
Only afterwards was the upper layer deposited and colonised by the crabs.
It seemed to them that there were two layers created at different times and not within a couple of days as Smit had suggested.
So this layer up here has nothing to do with the lower level, the two are separated by time, and by a considerable amount of time.
The next awkward fact was the existence of layers of muddy sedimentary rock, interspersed in the sandstone.
These they thought would have taken a long time to form.
They are made of fine grains which tend to build up slowly on the sea floor.
So these fine grains needed some time to settle, in fact an awful lot of time and in that case these sandstones would be separated by some time.
Then, near the bottom of the outcrop they discovered their third awkward fact - a layer of limestone.
Limestone is given away by its violent reaction to acid.
It can also take a very long time to form.
Finally, Keller and Stinnesbeck discovered their most awkward fact of all.
It lay eight metres below the outcrop.
It was another layer of spherules.
Chemical analysis suggested, these remains of tiny meteorites like the layer higher up could only have come from the Chicxulub crater.
For Keller and Stinnesbeck, this changed everything.
It was a Eureka moment, absolutely.
We danced, we jumped up and down and said we've got it.
They were excited because these spherules were eight metres below where any had been found before.
In geology, lower normally means further back in time, so for them it suggested an entirely different explanation for the whole rock formation and the extinction of the dinosaurs.
It was all to do with time.
After the Chicxulub impact, the spherules settled on the sea floor within minutes.
These were the spherules just discovered by Doctor Keller.
On top of the spherules there seemed to have been a period of normal sedimentation, eight metres of mud had built up.
According to Keller and Stinnesbeck this would have taken two hundred and seventy thousand years.
Next came the limestone layer, this too could have taken thousands of years to form.
Only then came the sand, washed in to form the sandstone.
Then those fine grained layers built up by sedimentation, that could have taken centuries.
Then worms apparently had months, maybe years in which to colonise some of the layers.
And only then, at the very top came the Iridium layer, which coincided with the extinction of the dinosaurs.
But by now Keller and Stinnesbeck calculated that three hundred thousand years had passed since the Chicxulub impact.
The Iridium must have come from an asteroid but it must have been a second impact, three hundred thousand years later for which we simply don't know where the crater is and it is that impact that probably finished off the dinosaurs.
If Keller was right, there hadn't been one impact but two.
The first, Chicxulub, would have been devastating enough It would have killed many dinosaurs but it did not drive them to extinction and it didn't create the Iridium layer.
According to Keller and Stinnesbeck, even as earth absorbed the Chicxulub impact, deep in outer space a second asteroid was hurtling toward the planet and this was the asteroid that left the telltale Iridium and finished off the dinosaurs.
At the end of 2001 an international team of scientists gathered near Chicxulub.
They used an oil rig to drill down over a kilometre into the heart of the crater.
They extracted hundreds of metres of rock cores, most of the rock had been pulverised by the Chicxulub impact but one section was undamaged.
It lay above the broken rock so it must have been created after the impact and just underneath the Iridium layer, so the question was how long had this section taken to form? If the mysterious rock had been formed slowly it would support Keller and Wolfgang Stinnesbeck.
What we can see here is the impact sediment which is all smashed off the rocks, we see bigger pieces and smaller pieces here, but then we find something completely different.
See this grey rock, this is a limestone and when I looked at the limestone with my head lens I noticed that the grains which are forming the limestone are very fine, so that would show to me that this sediment has been deposited under rather quiet conditions.
In other words, Stinnesbeck thought the rock had been formed slowly, by sedimentation.
There must have been a long time between the impact below the core and the arrival of the Iridium at the top.
Then they noticed something else in the core.
Traces of what looked like a green clay mineral.
Samples were sent for analysis by a technique called x-ray diffraction.
Every mineral has its own unique signature when exposed to x-rays.
The analysis revealed a clay mineral called Glauconite and Glauconite has one vital characteristic.
Glauconite needs many thousands of years to form, so it's indicating interruptions in the sedimentation of this limestone so it extends even more the time, the period of time which is needed for the deposition of this limestone.
Gerta Keller also had samples from the core.
She was looking for something quite different.
I just scanned them very rapidly and at first I didn't see anything, and then I began to see some shapes that looked familiar.
What Keller thought she had seen were microscopic plankton fossils called forams, but they were poorly preserved and it took an expert eye to make them out.
This is a 3D image of a perfect foraminifera.
This is what we found in the core, it is a cross section of the same species.
This is another 3D image, you see five chambers arranged in a circle, you can see the same chambers in the cross section.
Here is another species, each of the segments come to a pinched end and the same you can see on the cross section.
All these species went extinct at the same time as the dinosaurs, yet Keller had found their fossils in rock formed after the Chicxulub impact.
For Keller the meaning of the discovery was clear.
The fact that we find these species in the rocks above the Chicxulub impact rocks means that Chicxulub impact must predate the mass extinction by about three hundred thousand years and that just confirms what we already found.
Keller and Stinnesbeck thought they had won.
For them the evidence proved beyond doubt that Chicxulub was far too old to have caused the extinction of the dinosaurs.
But that was just one way to look at it.
When Jan Smit analysed the core, he saw something quite different.
He found evidence that the rock had been created quickly.
Look here for instance, you see here ripple marks, there's another layer of ripple marks, there are lines in the rocks, they all bear evidence for strong currents which implies Tsunami events.
Smit also examined the layers of green clay mineral.
He did his own x-ray diffraction analysis and found not Glauconite but a different mineral called Smectite.
Smectite can form quickly.
So they made another mistake, they misidentified the clay mineral with very severe implications.
And as for Gerta Keller's forams, Smit was in no doubt she had made another big mistake.
He took photographs through a microscope too - he found not a single fossil, just these inorganic crystals.
On these fuzzy pictures you can easily claim they are forams but I claim it's a fortuitous combination of those crystals in a very fuzzy picture.
If you take sharper pictures like these with a different method, you see they are totally different.
According to Smit, all the evidence suggested the core had been formed in just a few days by a Tsunami and that meant Chicxulub must be the killer of the dinosaurs.
This is still one and the same impact, we have the Chicxulub impact here, very quick sedimentation by Tsunami waves and at the very end we get the Iridium from the Chicxulub impact so the dinosaurs got extinct right here, as a consequence of the Chicxulub impact and not by a second impact on some unknown place in the world.
Jan Smit and Gerta Keller had fought each other to deadlock.
Smit still represents the view of the great majority of scientists in this field but there is now some support for Gerta Keller's position.
Neither side is about to back down.
We looked at all evidence and I refute everything.
There's not a single argument which holds ground.
He built his career on this impact Tsunami hypothesis and right now he's feeling this hypothesis crumble in front of his eyes.
But while two groups of scientists were obsessed with the debate about which was the guilty impact, another thought the whole thing was completely beside the point.
The objection was based on one simple observation earth has been hit many times and these impacts, even the really big ones, seldom coincide with mass extinctions.
For the very large craters that we think might have produced global effects, in some cases they don't line up with any extinction events whatsoever and in other cases extinction events, very well dated extinction events have occurred that are not associated with any impacts.
Because of this, some scientists had long doubted that any impact could have wiped out the dinosaurs and now there is new evidence which seems to reinforce those suspicions.
The new evidence relates to the forest fires that were said to have raged all over the world.
They seem to have been supported by soot in the KT boundary layers, but the problem with soot is that it can be blown thousands of miles from where the burning actually happened, so one researcher decided to investigate a more reliable kind of evidence charcoal! Large pieces of charcoal can't be transported great distances, they can't be picked up and lofted in smoke plumes or carried around the world so therefore if we can find charcoal at each KT boundary location we can be sure that burning occurred at each site.
In 2003, Claire Belcher scoured North America looking for charcoal.
She collected samples from eight KT boundary sites, from New Mexico in the south right up into western Canada.
In the lab she searched the samples for traces of charcoal, which show up white under the microscope.
She was amazed at what she discovered there was almost none.
What we did find however was a lot of this non charred plant material.
This was really surprising because it means that to have this amount of non charred plant material in the KT boundary rocks, there's no way that there can have been a globally extensive wild fire because it just simply wouldn't have left so much non charred material on the rock record.
Claire Belcher's results were consistent from Canada, all the way down to the southern United States, just two thousand kilometres from the crater.
Even here there was no evidence of burning.
If there was no burning across north America then how on earth could there be burning across the rest of the world because north America is so close to the impact site so basically it seemed there were no fires.
So the dinosaurs had not roasted.
A major predicted effect of the impact simply did not hold up.
And that raised a whole series of other questions.
What about the idea of acid rain, meant to have been as strong as battery acid, that should have destroyed the plants that many dinosaurs lived on and wiped out many sensitive animal species.
Biologist Dave Archibald has made a statistical study of the kinds of animals which went extinct and those that survived.
He found that many which should have died out because of acid rain but didn't.
This little tree frog like all amphibians is very susceptible to acid rain.
They have problems with their metabolism and finally they can't breed in the water.
If acid rain had occurred at the end of the Cretaceous, all amphibians should have disappeared.
They don't, almost all of them survived, as do all the turtles, the crocodiles, their cousins the alligators and most fishes.
This says that there can not have been battery acid strength acid rain at the end of the Cretaceous.
If there was no significant acid rain, it seemed another killing mechanism had been disproved.
And what about the so called impact winter? Dust was meant to have blocked out the sun that could have plunged the world into deep freeze for months or years, but again those little frogs got in the way.
Amphibians like all cold blooded creatures have problems dealing with sudden drops in temperature, especially if it lasts for many months or years.
If this had happened at the end of the Cretaceous, then many cold blooded organisms should have become extinct .
.
but they did not.
This tells us that there could not have been a long term drop in temperature.
So the dinosaurs did not freeze to death nor did they roast and nor did they die from the effects of acid rain.
There seemed to be so many holes in the impact theory, scientists began to dust down a more classical, evolutionary idea.
Maybe the dinosaurs had died out gradually and for completely different reasons.
Many scientists believe that about ten million years before the KT boundary, the dinosaurs' environment began to get markedly worse.
The lush coastal planes on which they depended began to turn arid, as sea levels fell dramatically.
Their natural habitat was disappearing.
They no longer have as much area to live in, they don't have as much food to eat, it's harder to get from one area to another.
Organisms are adapted for a certain type of environment and when those conditions disappear those organisms disappear.
Between ten million and three million years before the KT boundary, the dinosaurs seem to have suffered a dramatic decline.
One interpretation of the fossil record is that forty percent of dinosaur species went extinct.
And so we come up to about one to three million years before the end of the Cretaceous and we've got a group that seems to be vulnerable to any kind of extinction events that might be occurring at that time.
So it seems that by three million years before the impact, the dinosaurs were already in trouble.
The big question is what happened next? Did they stabilise or carry on dwindling? What is known is that their environment continued to worsen.
For about half a million years before the KT boundary, the world suffered one of its most destructive periods of volcanism ever.
Vast areas of land were ridden with eruptions.
A million cubic kilometres of lava spewed out in what are known as the Deccan Traps.
We're talking about the entire western part of India being a centre of volcanic activity with volcanoes and lava coming out, over a large area, over thousands of kilometres.
Volcanic greenhouse gases contributed to an apparently massive global warming.
Some data point to an average rise in temperature of eight degrees centigrade in the last half million years before the impact.
This would have been devastating.
We're talking about catastrophic effects in terms of changes in habitat, changes in rain fall patterns, changes in climate, all of these you can think of all of the things that are going on in the modern world magnified many many times, many many orders of magnitude indeed.
It is not clear what effect the Deccan Traps had on the dinosaurs.
The fossil record for this crucial period is difficult to interpret.
But the fossil record for other animals such as the ammonites is far better and using this some argue it's possible to infer what was happening to the dinosaurs.
Six million years prior to the KT boundary there were about twenty species of ammonites in the world's oceans.
Three million years before the KT boundary there were only fifteen or so and one million years prior to the KT boundary we have less than half of what we started out with, we have less than ten species so the extinction event has already been going on for millions of years.
The amazing thing is that we see the same pattern in the fish record, we see the same pattern in the terrestrial reptile record, we even see the same pattern in the mammal record.
All of these groups were undergoing an extinction event for millions of years and it would be absolutely amazing to me if dinosaurs weren't undergoing the same sort of extinction and indeed I think they were undergoing the same sort of long term extinction.
It seemed there was no role left for the asteroid impact or impacts at all.
The dinosaurs were doomed anyway: except there was still evidence of an environmental catastrophe at the KT boundary.
The awkward unavoidable fact that remains to support the impact theory was the fern spike.
The predominance of fern spores in the KT boundary layers suggests that all vegetation was destroyed for a time.
Those who are convinced that Chicxulub killed off the dinosaurs argue that the fern spike, caused by the impact, is still the key to understanding the mass extinction.
It's reasonable to think that all the standing vegetation all over the earth was destroyed, so we don't understand exactly how yet but certainly its evidence of a dramatic catastrophe in the terrestrial eco system from the impact.
Because the vegetation was not destroyed by wild fires, the most likely cause seems to be global darkness brought on by the impact dust and that would have been catastrophic for the dinosaurs.
Naturally if you're a plant eating dinosaur, all the plants that you're eating are dying so this of course is the end of you, and if you're a meat eating dinosaur preying on plant eating dinosaurs once they're gone, you're gone too.
The end of the dinosaurs now seems much more complicated than it once did.
It appears they were already dwindling under pressure from a worsening environment.
Massive volcanism and falling sea levels might have forced them to extinction, even if the asteroid had missed.
But there was a massive asteroid impact, either at Chicxulub or was it somewhere else entirely? And it does seem likely that this, at the very least finished off the remaining dinosaurs.
So it seems they were the victims of a lethal combination of circumstances.
Clearly dinosaurs were incredibly unlucky at the end of the Cretaceous to have all three things happen at the same time.