NOVA

DAVID ATTENBOROUGH: Dinosaurs.

Perhaps some of the most dramatic animals ever to have walked the Earth.

(dinosaurs lowing) They dominated the world for over 150 million years, until a huge asteroid struck the planet.

But how exactly did they die?

Paleontologists have been searching for the answer for decades.

And now new evidence is coming to light.

STEVE BRUSATTE: We're out looking for clues.

And each fossil is a clue, and that tells us something about what the world was like at that time.

ATTENBOROUGH: Since 2012, a team of paleontologists has been investigating a remarkable site deep in the Badlands of North Dakota.

The team's leader, Robert DePalma, hopes it holds evidence of what happened on the very last day of the dinosaurs.

(thunder rumbling) Could it even contain the remains of an animal that bore witness to that terrible event?

We've got all these bones in the ground right now, but the one thing that we would just dream of finding is that one dinosaur that died on the day of the impact.

BRUSATTE: The idea that there is a dinosaur fossil potentially that's a direct victim of that, that's very exciting.

MAN: Whoa!

DEPALMA: That's skin right there.

That's actually scaly skin.

My God!

ATTENBOROUGH: Can they find a dinosaur that died on the day the asteroid hit?

"Dinosaur Apocalypse: The Last Day," right now, on "NOVA."

ATTENBOROUGH: For ten years, paleontologist Robert DePalma and his team have been digging in a small corner of the Hell Creek Formation, an area famous for more than a century of major dinosaur discoveries.

They've already found a wealth of fossilized creatures in a patch of land they call Tanis.

(chuckling): Oh, wow.

ATTENBOROUGH: What appears to be a piece of fossilized skin from a triceratops; the unhatched egg and what looks like a pterosaur embryo; jawbones of a mammal called a pediomyid; and teeth and footprints of carnivorous dinosaurs like T. rex.

DEPALMA: There is no other dinosaur that has teeth like this.

ATTENBOROUGH: Many of these fossils were found in a thick layer of crumbly rock.

The rock here is really not quite rocky, and it just falls apart in your hands.

ATTENBOROUGH: Right above the crumbly rock is the K-Pg boundary, a layer of iridium-rich debris from the asteroid impact that hit the Earth 66 million years ago.

It marks the end of the age of dinosaurs.

CATHY PLESKO: If you look below this layer, you see fossils of dinosaurs.

If you look above this layer, no dinosaurs.

ATTENBOROUGH: The four-foot-thick layer of rock at Tanis is full of ejecta spherules.

DEPALMA: And beautiful-- look at that.

ATTENBOROUGH: Tiny glass droplets created in a major asteroid impact.

Robert thinks that this is compelling evidence that everything in the layer was buried while the spherules fell.

If he's right, and the spherules he's found can be matched to the asteroid impact, this dig site could provide a snapshot of what happened on the very last day of the dinosaurs.

BRUSATTE: Stories like this are eminently plausible.

Proving them is more challenging.

EMILY BAMFORTH: It opens up that whole debate about, how do we link catastrophic events to fossil and geologic deposits?

SONIA TIKOO: If we can both match spherules to the impact site, geochemically and in terms of radiometric ages, that's pretty accurate-- that's a smoking gun.

After ten years of digging, there is now enough evidence to piece together much of the story of Tanis and the creatures which lived here.

But how, exactly, did they die?

The asteroid that struck the Earth 66 million years ago created what is today known as the Chicxulub crater.

To find out if the ejecta spherules they've found in North Dakota can be linked to Chicxulub, Robert has come to the Diamond Light Source Synchrotron in the U.K.

Joining him is Phil Manning of the University of Manchester.

They've already run initial tests in America on over a dozen spherules found in different areas of the crumbly layer.

What have you found out so far?

These little glass spherules, these globs of molten material from the impact site, have a chemical signal that ties it with where they came from.

'Cause when an asteroid hits, it melts the ground that it hits, but also, that glass has a little bit of contamination from the asteroid itself.

And that gives you a unique geochemical fingerprint.

MANNING: We can see once we've scanned it, and looking at other sites from around the world-- Haiti, Mexico, and North Dakota-- we can get a baseline for what the ejecta should look like when it's related to the Chicxulub crater.

You can see each element here and the ratios of those elements.

And when we look at Tanis, it's a match-- I mean, it perfectly overlays.

So I think this is powerful evidence supporting that Tanis and Chicxulub are linked.

Wow.

And what do these findings mean for the rest of the fossils that you're finding in Tanis?

This data is key for the entire site, because once you have that link and you know what impact affected Tanis, then you essentially know that every object in that site, all the animals and the plants and everything buried in those sediments, are linked to the last day of the Cretaceous.

ANUSUYA CHINSAMY-TURAN: This is very important because it immediately gives a time stamp for the locality itself.

BAMFORTH: The Tanis site is like a window into a snapshot of time.

ATTENBOROUGH: With ejecta spherules found everywhere throughout the four-foot-thick deposit, Robert and his team seem to be able to link their site to a single day.

And the synchrotron here in the U.K. reveals something even more remarkable.

DEPALMA: So this is showing a beautiful synchrotron scan of the half of one spherule.

The glass is a good geochemical fingerprint, but, when we look at the entire thing, we see something quite unexpected.

That's your entire spherule.

What's this?

In this, we've got a little bit of a nugget.

There was a little particle right there.

So we scan it, and that's a lot of iron in there.

Over here, we've got chromium, a big peak in chromium.

Over here, we've got a big peak in nickel.

And the abundances of iron, nickel, and chromium, all together, that matches what you expect to see in a meteoric body.

That does not match what you would normally have down here.

So this is extraterrestrial material.

If you were to sort of grind up and stuff into a, a spherule a piece of meteorite, that's what it's gonna look like.

DEPALMA: This could be piece of the Chicxulub asteroid.

The piece of the bullet that killed the dinosaurs.

No.

Robert's team may have found a fragment of the asteroid itself in North Dakota, physical evidence linking this site to the Chicxulub impact.

But Tanis is almost 2,000 miles away from where the asteroid hit.

So exactly how did the asteroid cause the death of the animals here?

To answer that question, Robert is searching in something he calls the mass death assemblage.

DEPALMA: Right here we've got this intertangled mass of fish.

There's one fish here, another sturgeon goes this way, underneath the body of a paddlefish.

There's another sturgeon that goes this way, underneath this log, and continues out the other side.

And his head hit that log, and has deflected downward at a 90-degree angle.

ATTENBOROUGH: Robert uncovered a tangled mass of fossilized creatures and logs surrounded by spherules and crushed together in what's known as a logjam.

He has a theory that the creatures were swept to their death in some kind of turbulent surge of water, and quickly entombed in sediment, which is why they are so well-preserved.

But what could have caused the wave?

One hypothesis is a tsunami: the asteroid hit at sea.

We talk about a tsunami of completely different scale, much higher, much larger than we've ever seen before in modern tsunamis.

SEAN GULICK: So if you had 6,000 feet of water, at least half of that would have left as the rim wave.

So at least 3,000 feet high at a minimum.

ATTENBOROUGH: The tsunami raced towards land.

TITOV: When they reached the coastlines, they were still very high waves of up to 300 feet high.

At least-- probably as high as 1,000-feet-high wave.

That's, that's very impressive wave.

Imagine the wave that's the size of a building approaching the coastline.

ATTENBOROUGH: In the Late Cretaceous, North America was divided by a narrow sea that has been called the Western Interior Seaway.

The tsunami could have theoretically traveled up this towards Hell Creek.

Tsunamis generally travel at about the speed of a jet plane.

It's not something you could, say, run away from.

It had plenty of energy to get over the, the coastline.

It could easily still have been tens of meters high by the time it reached well into the seaway.

(wind blowing) ATTENBOROUGH: Could the rapid deposition at Tanis have been caused by a tsunami?

To test the idea, the team needs to look at the timing.

DEPALMA: Oh, which fish is that?

It's a new, it's a new contact.

New one, yeah.

ATTENBOROUGH: If a tsunami buried the fish, it would have to have hit while the ejecta spherules were falling, because spherules were found everywhere, including in the fishes' gills.

(typing) So much depends on determining when the spherules were falling at the site.

RILEY BLACK: Modeling the ejecta always has error bars on it, in that we're not there to measure it and we've had no equivalent impact like this on Earth since then.

PLESKO: But we can look at the computational models that we do and say, all right, this material is coming from this point, it's now moving away this fast, with about this much mass.

And then we can tell, with the sorts of equations that we might use to calculate the trajectory of a cannonball, where it would go.

And we can observe from these simulations how long it takes these ejecta to reach their final destinations down to the order of a few minutes.

ATTENBOROUGH: What the calculation shows is surprising.

Robert and his team have found that these ejecta spherules landed at Tanis between 13 minutes and two hours after the impact.

So, if a wave buried the fish, it must also have reached the site within two hours.

Data from recent tsunamis show even a powerful wave would take much longer than this to travel almost 2,000 miles from the impact site to North Dakota.

So, if it wasn't a tsunami, what could have caused a surge of water at Tanis?

Stein Bondevik is an expert in tsunamis.

(water sloshing) BONDEVIK: The fjords in Norway are very special.

We have tall mountains surrounding bodies of water.

So the water is usually very calm.

ATTENBOROUGH: In 2011, something very strange happened.

The water in the fjord began to move violently.

BONDEVIK: The height of the water increased by one-and-a-half meter, like a maelstrom with the turbulent water.

Someone said that the fjord was boiling.

ATTENBOROUGH: News started to roll in, there'd been an earthquake 5,000 miles away in Japan.

BONDEVIK: A journalist from the local newspaper called me and he said that people were observing waves here in the fjords.

I got a video clip of the waves.

And I saw immediately that they looked like a tsunami wave.

Here you can see that that the fjord is perfectly calm.

(water rushing) But at the beach here, you could see that the water is sloshing back and forth.

And no one had ever seen anything like it.

Some people got very upset and afraid.

ATTENBOROUGH: A magnitude 9.0 earthquake had devastated the northeast of Japan.

But how did that affect a fjord so far away?

BONDEVIK: So no one in Norway could feel the earthquake.

but I could see that the times matched the arrival of the waves here in the fjord.

ATTENBOROUGH: Eventually, Stein and his team realized that this might have something to do with seismic waves, shock waves that pass quickly through the earth during an earthquake.

BONDEVIK: So it took only 12 minutes before the first signal of the earthquake in Japan reached all the way here to, to western Norway.

(birds twittering) ATTENBOROUGH: So it was the seismic waves that caused the normally calm water in the fjord to slosh turbulently back and forth.

BONDEVIK: Just thinking of that, scientifically, it's, it's fantastic.

ATTENBOROUGH: Could something similar have happened in Tanis?

MARK RICHARDS: Been large weather fronts coming through... (indistinct chatter) ATTENBOROUGH: Geophysicist Mark Richards has been studying the site for several years.

The events in Norway support a hypothesis that he's been working on with Robert's team about what could have caused the surge of water here.

RICHARDS: A tsunami can't get here in less than minimum 12 hours.

But seismic waves traveling from the Yucatan impact site to North Dakota can arrive here fairly quickly.

ATTENBOROUGH: In the Late Cretaceous, the Western Interior Seaway that bisected North America could have been connected to Tanis through the extensive river system that once flowed here.

RICHARDS: If you have a very large body of water, like the Western Interior Seaway, and you can shake it back and forth, you can generate a large water wave coming up this river at Tanis.

GULICK: So this is bigger than any tectonic generated earthquake.

You would have shaking literally everywhere on the planet.

ATTENBOROUGH: So, their hypothesis suggests seismic waves from the impact could have caused surges of water in the Hell Creek river system.

RICHARDS: Seismic waves get here quickly enough to cause this wall of water, coming up the Tanis river, inundating this area, arriving at the same time these spherules are still falling out of the air.

ATTENBOROUGH: If they're right, seismic waves traveling through the earth, could have caused a powerful surge of water at Tanis... (wave crashing) ...at the same time as spherules fell.

And ultimately dumping it on the Tanis sandbank, burying everything in the churned-up mud.

Debris and fine iridium dust from the asteroid would have gradually covered the deposit, forming the K-Pg boundary.

Over millions of years, the surge of mud would become the four-foot-deep layer of crumbly rock.

RICHARDS: And that's the beauty of Tanis.

What you're seeing is a deposit that is literally recording the last, say, 45 minutes to an hour-and-a-half of the Cretaceous.

ATTENBOROUGH: If the extinction of the dinosaurs was a crime, the detective solving it would have plenty of evidence.

They would see that the asteroid was in the right place at the right time.

They would see that no dinosaurs survived after the hit.

They would have a piece of the murder weapon, a fragment of the asteroid.

But they would be missing one very important thing: a body.

BLACK: A lot of the bones that exist from those last Cretaceous days were basically destroyed.

BRUSATTE: As far as we know, we've never actually found a fossil of a dinosaur individual, a single skeleton, let's say, that we can unequivocally say was there on the day the asteroid hit.

ATTENBOROUGH: But before the site was timestamped to the Chicxulub impact, Robert's team did find part of a triceratops in the crumbly layer at Tanis.

So could that be the body?

A dinosaur that died on that day?

Something that would help them would be establishing the cause of death, which can be difficult when you only have a piece of skin and horn to go on.

This is the horn after they've cleaned it up.

The team is particularly interested in these lines here.

And they found that the fractures go right through the horn.

So rather than dying as a result of the impact, they wondered whether it had been killed in a fight.

(dinosaurs growling, horns clashing) But when they looked at the fractures in more detail, they found signs of new bone growth here.

An indication that the bone had started to heal.

So it looked as though the triceratops survived the event that broke its horn.

(dinosaur grunting) Could this triceratops have survived until the day of the impact?

This drooping in the skin and the disarticulation of some of the bones suggested to the team that there was decay underneath.

That means its body had started to rot before it was entombed and preserved by the surge.

So, it seems that this dinosaur didn't die as a result of the asteroid impact.

(dinosaur grunting) Given the signs of partial decay... it's likely this Triceratops wouldn't have lived to see the last day of the dinosaurs.

(thunder rumbling) However, the triceratops fossil does show that dinosaurs were alive shortly before the asteroid hit, perhaps even within weeks of the impact.

This is an extraordinary discovery, and one that has never been found before.

But if it's true that dinosaurs were here until the final weeks before the impact, there could be even more still to find in this deposit.

DEPALMA: This is, like, looking down onto the side of a dinosaur that died weeks to months before the impact.

That is such a cool thing.

We've got all these bones in the ground right now, but the one thing that we would just dream of finding is that one dinosaur that died on the day of the impact.

ATTENBOROUGH: And the weather isn't helping his search.

Got... (groans) That therapod print is toasted.

BURNHAM: Yeah, it was in a low corner.

Look, it's all... it's full of mud and water.

The problem is it's wet, look.

See, if we're not careful, we're gonna lose the print.

And that's the biggest theropod print we've got.

I see some areas that could use glue right now, too.

ATTENBOROUGH: The team is racing to excavate dozens of fossils before the rains wash them away.

DEPALMA: We are up against the clock here.

This stuff that could be exposed right now is gonna get ruined by the rain.

ATTENBOROUGH: But then, the team comes across something that looks very unusual.

(speaking indistinctly) BURNHAM: What is going on right there?

Are we sure this isn't crocodilian?

That's not crocodilian.

No.

GURCHE: Let me try this piece right here.

I'll go in from the top and then twist up, and it separates right on that line.

DEPALMA: Oh... that's skin right there.

That's actually scaly skin.

Oh my God.

No, no, no, no, no, look, look, look.

Look at that pattern right there.

Have you ever seen elongated scales like that before, Dave?

BURNHAM: Scutellates in birds.

Just careful.

Oh, my God.

BURNHAM: It's changing again.

It's changing again.

My God.

We're seeing it for the first time in 66 million years.

I think we got ourselves a dinosaur.

ATTENBOROUGH: A dinosaur fossil.

In the same mass death assemblage as the fish with the spherules in their gills.

DEPALMA: This is the most incredible thing that we could possibly imagine here, the best-case scenario.

We're excavating this mass death layer of fish from the surge sent up by the impact, and we've got dinosaur remains.

The one thing that we would always want to find at this site, and here we've got it.

This is unreal.

I, I, I cannot process this in my brain.

No, I am absolutely blown away by this.

Just my heart is literally pumping out of my chest wondering what is behind there, just a couple of centimeters back in the outcrop.

What is waiting for us back there?

This is amazing.

ATTENBOROUGH: The team keeps digging.

BURNHAM: So this could be a ribcage, it could be laying against ribs that are curved.

DEPALMA: There's something here.

That's hard.

That's bone right next to the skin.

Look, that, that's an articular surface right there.

So this is either a hip or a shoulder element.

ATTENBOROUGH: After hours of painstaking work... DEPALMA: And we can go from the thigh of the animal.

There's the knee.

And then you've got the little calf muscles of the, uh, dinosaur over there bulging out, and you go down to the anklebones, and these are the toes of the feet.

We've got nails at the tips of the toes.

It's a beautifully preserved leg, all articulated, covered with skin.

ATTENBOROUGH: The complete leg of a dinosaur.

DEPALMA: In my wildest dreams, I never expected to find a dinosaur leg in this deposit.

Yeah.

I mean, and then it's got skin and tissue.

It does look just like a drumstick.

It looks like a Thanksgiving turkey just laid out in the ground.

ATTENBOROUGH: Robert and his team think they've found the body missing from the crime scene-- a dinosaur that might itself have witnessed the cataclysmic impact.

DEPALMA: Dinosaur fossils are not known from the last years of the Cretaceous.

And it was unclear whether they were already extinct, or in decline, or what was going on.

So they were just sort of absent.

And this answers that question: were dinosaurs still there then?

Well, yes, they were there weeks to months before the impact.

This one likely died in that surge.

ATTENBOROUGH: But such big claims need verification.

RILEY BLACK: Paleontologists, we kind of fight like tyrannosaurs.

There are lots of different opinions.

There are lots of different hypotheses.

And this is all because science is a process.

Science isn't a matter of just going out into the rock, and we found a new dinosaur.

STEVE BRUSATTE: So I think we always want to be skeptical, but I think we want to be extra skeptical when there are big claims.

And it all goes back to that expression that we hear all the time, and that's that extraordinary claims require extraordinary evidence.

ATTENBOROUGH: Robert is in the process of sharing the team's finds with the wider scientific community.

As part of this process, he has brought the dinosaur leg to London, England, to get a second opinion from paleontologist Paul Barrett-- an expert in ornithischian dinosaurs from the Natural History Museum.

So what do you think this might be?

When we look at the leg, it has claws like the claws we see in small, agile, bipedal running dinosaurs that are plant eaters.

We can also rule out things like Triceratops partly just because it's not big and stocky.

And the proportions of those legs are also different from some of the other plant eaters we see in that they have this rather long ankle and shin compared with its thigh bone.

So as we narrow those possibilities down what we're left with probably is an animal called a thescelosaur.

(dinosaur grunts) (growls) ATTENBOROUGH: Thescelosaurus are thought to have lived next to rivers.

(dinosaur grunts) They had leaf-shaped teeth common amongst herbivores, and claws on their short front limbs, which they may have used for digging.

At the front of their mouth they had specialized, pointed teeth that could help them to pull roots out of the ground.

So it's possible they dug for food.

(biting flesh) But how did the thescelosaur that Robert's team found die?

Could it have killed by another dinosaur?

BARRETT: It's a possibility.

This is a relatively agile animal.

And that turn of speed would've been its primary defense against the large predators living alongside it.

(insects chirping, dinosaur grunting) DAVID MARTILL: Whenever we're excavating a dinosaur, one of the things that we're always keen to know is how did the animal die?

It's not always easy to do that.

BRUSATTE: So maybe we can find evidence for things like broken bones that didn't heal back up.

Sometimes we can even see things like bone tumors and gout.

CHINSAMY-TURAN: There are some wonderful fossils where you can find bite marks on them.

You can even find a, a predator tooth buried within the bones.

EMILY BAMFORTH: In science, we don't prove things.

We just disprove some things.

Generally speaking, unless a cause of death leaves a signature on the skeleton, it's hard to tell.

ATTENBOROUGH: CT scans Robert and the team have taken of the dinosaur leg allow a closer look at what the animal might have gone through before it died.

It doesn't seem to me like there is any evidence that this animal was predated.

None of the obvious tooth marks or left over bits of carnivore teeth to suggest it's been eaten.

So we can see that the bones look okay.

So this was an animal that was probably living and healthy at the time that this happened to it.

ATTENBOROUGH (voiceover): Could this be a dinosaur that was swept up in the surge?

BRUSATTE: The idea that there is a dinosaur fossil, potentially, that's a direct victim of that, that's very exciting.

I think ultimately it comes down to a couple of things.

You know, are there injuries on that fossil that show that this dinosaur was bobbing about, heaving about in the water?

So are there things like breaks on the bone or other things that have not healed?

This is actually a shoulder bone.

And this bone in the living animal would actually be way over here.

And similarly this little bone here would have been from about a third of the way along the tail, maybe halfway down.

So somehow these two bones have been telescoped together.

Yeah.

So maybe this animal has been tumbled around.

Could this be a victim of the meteor strike?

I think it's entirely possible.

We've ruled out a lot of other possible causes of death for this animal.

So it could well be that this was an animal that was there being tumbled around in its death throes in that river as a result of the asteroid impact.

Paleontologists do depend a lot on tragedy.

Every little disaster is the material we need to actually develop our subject.

Tragic for the individual concerned, but we're just really happy that it happened.

ATTENBOROUGH: After years of work at this site, Robert and his team have uncovered unprecedented detail about the animals living there.

And he thinks that many of them were alive on that fateful day when the asteroid devastated our planet.

But how exactly did they die?

The team's finds give us new clues to answer that question.

One of the most important days in Earth's history probably started much like any other late spring morning.

We think it was late spring because paleobotanists have found key evidence about the season from fossilized flowers.

The Tanis finds are consistent with this, including the fossils of young fish that died at the size they reach at that time of year.

Perhaps this day, that would end with so much death, began with something different.

A new life.

Robert only found one pterosaur egg, so it's possible that it had a brother or sister that hatched before the impact.

(pterosaur squawks) To see the world for one final day.

(pterosaur squawks) (insects chirping, creature bellowing in distance) No one can be certain of the exact timings of the day when the asteroid collided with our planet.

But it's estimated that within just 40 minutes of the impact, the consequences for the creatures of Tanis would have been profound.

Based on the team's finds, and the latest evidence from other scientists, this is how the catastrophe might have unfolded.

The asteroid is around seven miles across-- bigger than Mount Everest-- and traveling at close to 45,000 miles an hour.

The impact causes an explosion with over a billion times the power of the first atomic bomb.

GULICK: : It comes in so fast that it wouldn't even have been visible passing through the atmosphere.

Right, it would have just come in and hit in a moment.

ATTENBOROUGH: At Tanis, almost 2,000 miles away, there might have been an initial flash of light, yet it is completely silent.

But at the impact site... ...the asteroid vaporizes.

More than three trillion tons of rock are ejected into space in a blast of superheated violence.

(rocks crashing, wind whipping) Winds higher than 600 miles a hour.

A colossal earthquake.

Followed by a ring of massive tsunamis.

All the while the creatures at Tanis go about their business... (pterosaurs grunting) ...just like any other day.

(animals chittering) (dinosaur makes guttural sound) (snorts) (thunder rumbling) (pterosaur squawks) The evidence suggests that baby pterosaurs may have emerged from eggs ready to fend for themselves.

And that includes... Flying?

Well, almost.

(pterosaur chittering) Elsewhere, as the reverberations of the impact race out across North America... (low-pitched rumbling) ...dinosaurs and creatures of all shapes and sizes are obliterated by the blast... Incinerated in a firestorm unlike anything seen since.

TIKOO: If I were a dinosaur standing on the coast of North America, I would just see a flash and a fireball coming at me and then I would be fried.

All you feel is an awfully sharp, stabbing pain in your ears, then you explode.

ATTENBOROUGH: At Tanis, for a few more precious minutes, life continues.

But the clock is ticking.

(creature chittering) (dinosaurs growling, horns clashing) The blast from the impact never reaches Tanis, but seismic shockwaves do.

(creatures chittering) (rumbling) They are far more powerful than any earthquake ever recorded.

PLESKO: If you were standing on the Gulf Coast of Texas, that magnitude 12 earthquake would have been strong enough to actually jam your femurs up into your body cavity ATTENBOROUGH: While the earthquake that reached Tanis was likely less destructive, the effects would have been felt by all that lived there.

(loud rumbling) Seismic waves are now slowly shaking the whole region, causing water to slosh and churn.

At Tanis, these strange currents in the river may be some of the first signs of what is coming.

(thunder rolls) Next, it begins to rain.

(pattering) Ejecta spherules are falling back to earth.

As the spherules plummet, friction heats them until they're red hot.

They soon transfer their heat to the atmosphere... which grows hotter by the second.

(sizzling) As the searing heat builds, the creatures of Tanis are fighting for their lives.

(dinosaur bellowing) And then, as seismic waves rock the whole region... ...a violent surge wave 30 feet high rushes up the Tanis River.

Surviving the turbulence of the surge is a challenge even for the best swimmers.

Then, the slow but powerful rocking of the river system draws the water back.

A large, robust animal like a T. rex might have survived the surge.

(pattering) As might a hard-shelled reptile.

But there is much more to come.

As billions of tons of superheated spherules continue to fall, the atmosphere gets even hotter... ...igniting dead leaves and sparking wildfires.

Earthquakes... Fire... Devastation.

(dinosaur grunts) Little would survive for long, on land... (bellowing) Or in the air... (pterosaur groans) BLACK: The air around the planet was effectively set to broil.

This was something that you couldn't escape if you were out on the surface.

ATTENBOROUGH: Those that live underground may have had a better chance.

(creature chittering) As the slow rocking of the river system continues to move the water to and fro, another powerful surge hits the riverbank.

(loud crashing) For most, there is no escaping the destruction.

For many of the creatures here, their stories end underwater.

No one knows if the giant tsunami caused by the impact ever reaches this far north, but life here has already changed forever.

The mud the two waves leave behind will gradually turn into the thick layer of crumbly rock entombing the creatures which died here.

(wind whipping) Until 66 million years later, when they are finally unearthed.

CLAYTOR: We have a general idea of what horrors were unleashed on the landscape by the asteroid impact.

But I think these sites may give us the ability to actually put them in sequence and understand exactly what these organisms went through.

Even though there is a lot of debate, and there is a lot of controversy, every new thing that we find, every new hypothesis that's put forward, whether it's accepted or rejected, gets us a little bit closer to doing that mental time travel and imagining ourselves back in that Cretaceous world.

(fire crackling) ATTENBOROUGH: Robert's finds have helped us understand, in remarkable detail, what might have happened at Tanis in the minutes after the asteroid impact.

But what about the rest of the world?

Fires rage, destroying many of the world's forests.

As the horrific day draws to a close, many of the world's dinosaurs are likely already dead.

Research shows that the angle at which the asteroid hit, and the sulfur-rich rocks at the impact site, amplified the devastation.

Without sunlight, most plants died, and food became scarce.

(thunder rumbling) As the weeks and months passed, any dinosaur left alive would have died of hunger.

In the oceans, it was the same.

Nearly all of the world's plankton died, leading to the starvation of most marine creatures.

It's thought that the impact winter that followed caused a global temperature drop of at least 48 degrees Fahrenheit.

After this huge change in climate, the fossil record tells us that three-quarters of all species, including the dinosaurs, were wiped out.

(crackling) PLESKO: The location of the Chicxulub impact really was a worst-case scenario.

If the asteroid actually had come in, you know, 30 seconds earlier, 30 seconds later, it would have actually hit the Atlantic Ocean or the Pacific Ocean and not the sediment-rich, sulfur-rich target of the Yucatan Peninsula.

Forests collapsed.

The plant eaters didn't have any food to eat; they died.

The meat eaters didn't have any plant eaters to eat; they died.

Ecosystems collapsed like houses of cards.

BLACK: This unintentional accident that just was set in motion long before dinosaurs even existed.

And it just happened to be the one case of bad luck, the one worst day in the history of the planet.

ATTENBOROUGH: Studies suggest that the planet was in semi-darkness for around a decade as dust and soot slowly fell to Earth.

But then came something wonderful... A new beginning.

Once the dust cleared from the atmosphere, and the sunlight returned, plant life was gradually restored, led by ferns, the spores of which had lain dormant deep underground.

And the world began to turn green once more.

But what about the animals?

One of the reasons some mammals survived the Great Extinction were burrows.

During the impact winter, a burrow would have provided warmth, protection, and a place to store food.

(sniffing) Mammals which were able to thrive in the aftermath were resourceful omnivores.

And the insects which survived could have been one source of food.

Their size would have been another advantage.

(chittering) When catastrophe strikes and food is scarce, the largest tend to die out, whilst the smallest often survive.

(creature chittering) And they weren't alone.

The turtle found at the dig site may have been unlucky, but many others survived.

As did crocodiles, snakes, and many fish species.

BRUSATTE: Life has found a way, and life is now thriving again, and it is those ecosystems formed in the recovery from the asteroid that are the foundations of our ecosystems today.

It's kind of amazing that we're able to put our finger on this one line in the rock.

And as much as we miss the dinosaurs, say if this hadn't happened, we wouldn't be here.

ATTENBOROUGH: And as for the dinosaurs... did the impact really kill them all?

Well, this beautiful fossilized feather isn't from a bird, but from a predatory dinosaur.

So, we have to be careful when we say that dinosaurs are extinct.

Because what we call birds originally evolved from the smallest feathered dinosaurs.

So, to be correct, we should say all non-avian dinosaurs are extinct.

The finds from Robert and his team have given us a more detailed picture... (insects chirping) ...about what might have happened on the day that destroyed the largest beasts ever to walk the Earth.

Dinosaurs were perhaps some of nature's most extraordinary creatures, dominating the planet for over 150 million years before they became extinct.

(dinosaurs grunting) But extinction comes in different forms, and many of the amazing creatures and plants alive today are also threatened.

It's possible that humanity is having as big an impact on the world as the asteroid that ended the age of the dinosaurs.

As human beings, we are unique in our ability to learn from the distant past.

The question is: will we use that ability wisely and do our very best to protect the millions of species for whom, alongside us, this planet is home?

(water spraying) (waves crashing)