Horizon (1964–…): Season 39, Episode 13 - Earthquake Storms - full transcript

Four years ago the lives of these Turkish firefighters changed for ever

An earthquake of catastrophic size tore their city apart

Even after 72 hours of non-stop working day and night without any sleep

when we found someone alive it gave us such a morale

boost that we forgot everything

Although thousands died

many survivors owed their lives to the actions of these men and women

People were helpless and as the Fire Brigade

we were the only ones they could turn to

but there were times when even we felt helpless

I remember having to hide and cry secretly many times



But now a frightening forecast suggests it is only a matter of time

before these fire-fighters are needed again

Scientists have identified a new phenomenon called an earthquake storm

and one is blowing across Turkey

If their forecast is correct then next in line is a city of 14 million people

On August 17th 1999

at just after three in the morning

a massive earthquake struck the Turkish city of Izmit

Thousands of people fled as their homes collapsed around them

I couldn't fully understand what was going on

I didn't realise it was an earthquake as I wasn't fully awake

Mum was crying and she said ?m?r,the house is going to collapse

and I replied don't worry Mum

just hold me tight and that was the last conversation we ever had



The earthquake claimed the lives of some 25,000 people

The destruction spread over hundreds of kilometres

and much of the area was reduced to rubble

People were so shocked

Buildings had collapsed

Some of them survived,but some were still under the rubble

Some of them managed to get out of their home

and they just stood there completely dumbfounded

They didn't know what to do and could not think of

ways to help the trapped victims

Everyone was in shock

For the people of Izmit the disaster was totally unexpected

but what they didn't know was that a year earlier

the entire disaster had been forecast

We made the calculations, we knew that Izmit was dangerous

Our work showed that this area was at risk

an earthquake would come and that tens of thousands of people were likely to die

And yet that warning was not heeded

Few preparations were made,few contingency plans were put into place

When the Izmit earthquake happens

there are two sorts of feelings it sort of arous

arose in me. One was frustration

The other one was anger,anger with myself

anger with others because sort of

we knew it was coming and yet nothing had been done

Now the same scientists who highlighted the threat facing Izmit

have made another forecast

If they're right the danger could be heading for another city

a city much bigger than Izmit,a city with over 14 million people:

Istanbul

because scientists believe that Turkey is in the middle of a new phenomenon

where one earthquake triggers another and then another

Some are calling this an earthquake storm

The trail of clues that led to the discovery

of the earthquake storm began here in Crete

2,000 years ago the Greek island was one of the key

trading centres in the Mediterranean

but it was shattered

when the island was apparently struck

by the single biggest earthquake in recorded history

Evidence for that huge event is everywhere

This line here separating the light rock

from the dark rock is the ancient shoreline

and it carries on along this coast at the same level

Years ago the sea would have been up here lapping against

the rocks here and then something dramatic happened

One day the land suddenly rose up by about 6m here and the sea just dropped

A drop of six metres is the mark of a truly great disaste

A sudden jump of a coastline by this size really is

a telltale signs of a massive earthquake

Archaeological evidence showed just how devastating

this one event apparently was

With an earthquake of this size it just crushes towns in seconds

When the earthquake starts to stray the first few things

the walls start to go,the roof starts to come down

and anyone that's underneath is just crushed

What we're looking at here is crushed skeletons lying underneath masonry

and roof tiles and,and in amongst this all we've got columns fallen down

we've got pottery,everything else.This is everyday life

just crushed by an earthquake

And amongst the remains lay a vital clue as to when the earthquake took place.

This particular coin is dated to Contantius II

and that puts it roughly about 360AD.

Now this coin was found in the skeletons of the people crushed under the ruins

and what that means is the earthquake destruction

is dated to shortly after360AD

Ancient writings provided more evidence for the disaster

and dated it precisely to July 365AD

and for centuries many scholars found signs that the destruction

from the 365 event had spread far wider than just Crete

Terrible disasters took place all of a sudden throughout the world

Shortly after first light preceded by heavy and repeated thunder

and lightening the whole stability of the Earth was shaken

As more historical evidence was discovered around the Mediterranean

more damage was attributed to this one huge event

Cities 1500km apart appeared to have been struck by the same earthquake

A destructed Roman villa in Sicily

a destroyed Roman city in Libya,a destroyed Roman city in Cyprus

destroyed buildings in Alexandria,all of these have been put together

to generate this just ginormous earthquake that struck the length

and breadth of the Mediterranean

It seemed that few places in the Mediterranean escaped this destruction

Its range was so enormous this one earthquake became known as

the Universal Event

This was a truly giant earthquake

In fact,in terms of human history this was the biggest earthquake

the Mediterranean had ever seen

But despite all the evidence for this one massive earthquake

there were some who had their doubts

It was just too big

Historian Emanuela Guidoboni has studied ancient earthquakes

for decades and was troubled by the idea of this Universal Event

The area of damage was enormous

but it just didn't add up because the damage area extended

from Cyprus to Sicily,North Africa,Crete,reaching as far as Southern Italy

It simply wasn't credible

Intrigued Emanuela re-examined some of the historical texts

She was drawn to one writing in particular

The scholar Libanius wrote an epitaph for the death of

the Roman Emperor Julian the Apostate

In this epitaph for Julian he depicts these great images of earthquakes

that shake the land and gives examples of Libya,Sicily,Greece

Like a horse tossing its riders she has destroyed a great number of cities

many in Palestine and all those in Libya

The greatest cities of Sicily lie in ruins

as does every city in Greece except one

Many historians believe the epitaph referred to the 365 event

but Emanuela found something that didn't quite fit

The dating of Libanius's text is fundamental for

understanding the earthquake of 365

Libanius wrote the epitaph for Julian on the death of the Emperor

That is in May 363

In other words,Libanius had written his epitaph two years

before the Universal Event

and there were other writings that didn't make sense

It seemed that other earthquakes had been misdated

There had been seven earthquakes before 365 and three afterwards

so in 12 years there were 11 earthquakes

almost one earthquake a year

There was a seismic crisis in the Mediterranean

For Emanuela the theory that one single earthquake had wiped out

most of the Mediterranean in 365 had to be wrong.

There had been no Universal Event

Instead,it appeared that a cluster of separate earthquakes

had exploded across the region in a storm

each event seemingly triggering the next

It was an ingenious idea,but the suggestion that one earthquake

could set off another had always been regarded as scientific nonsense

Earthquakes are one of the most studied natural phenomenon of all

In the last 100 years they've claimed the lives of over

one million people and many more have been left homeless

Although extremely complex,scientists are beginning

to understand what causes these events

They know that most earthquakes strike where the Earth's

great continental plates push against each other at fault lines like this

They also understand how they happen

Well this is an earthquake machine

and we know the Earth to be composed of 12

or so plates and the plates are in constant motion,which is represented

here by the winch,and along the boundaries of all of

these plates are faults and this is a fault where the sandpaper

meets the brick in frictional contact

and the rubberiness of the Earth,represented by this elastic cord

stores the stress that builds up along the boundaries of these plates

and eventually the stress overcomes the frictional resistance

and we have an earthquake

But these faults stretch for thousands of kilometres

Pinpointing precisely where an earthquake will strike has always been a mystery

We know there's going to be big earthquakes over time

We just don't know exactly when they're going to happen

They could happen 10 minutes from now,they could happen 10 years from now

There's just no way to predict it

The one thing they did know was that smaller tremors called aftershocks always

followed a main event, but finding a connection between large earthquakes

years apart had proved impossible

Scientists had to accept that the planet's most destructive

tremors struck randomly and were totally unpredictable

Basically I'm throwing darts at a map and so the whole world of earthquakes

was viewed as dart throws that were independent in which nothing about the last

earthquake was going to condition my ability to say

very much about the next earthquake.

So the idea that earthquakes were random became the accepted view.

The trouble was the historical evidence suggested that

large earthquakes were very much connected.

We're finding these earthquake clusters at other times in human history

We've got them in the 18th century and the 15th century in Turkey

and people are even suggesting that the end of the Bronze Age civilisation

across the whole Mediterranean region could have been due to one

big earthquake cluster,so the big question is:

is that just random is that just chance

or we actually get a process where one is triggering

the other which is triggering the other

The archaeological evidence was suggesting one thing

but the scientists could just not explain it.There was simply not enough evidence

To understand how earthquakes might interact they needed

far more data than the historical records could provide

There was one obvious place to look:

the San Andreas fault in California.

One of the most active earthquake zones in the world this great scar

on the Earth's crust was created when one continental plate clashed with another

Seismologist Sue Hough has been fascinated by thus area all her life

I've been driving all over California looking for places like this

places where you can see evidence of faults in the ground and what's remarkable

about this place is that usually when plates slide past each other the

effect on the landscape's very subtle and here you can stand not only in

a fault zone,but actually on a plate boundary and that's really remarkable

You can go a few steps over here and you run into the Pacific plate

and that continues all the way to Japan almost

A few steps in this direction and you run into the North American

plate which keeps going all the way to New York

so you have massive tectonic plates on either side and this is

the one place where they meet and this is

the place that great earthquakes are going to happen

What makes the San Andreas so notorious is that sitting

at one end is San Francisco Bay

The six million who live here know an earthquake could happen at any time

Famously one struck the city in 1906 killing over 700 people

Another hit in 1989.This time over 200 people lost their lives

The earthquake hit at the start of San Francisco's rush-hour and on a key double

deck motorway in Oakland the top tier collapsed trapping hundreds of cars

on the lower tier on a mile-long stretch

With so many lives at stake

understanding how earthquakes behave along the fault is critical

This has made the San Andreas one of the most studied faults in the world

and along it there is one place more closely observed than anywhere else:

Parkfield, the earthquake capital of America

Parkfield is wired with a huge network of high-tech equipment

John Langbein is one of those in charge of these

sophisticated earthquake monitoring device

Here at Parkfield we have a variety of different instrumentation

some of which include seismometers which measure

vertical and the horizontal motion

This is a GPS unit.This is a two-colour EDM

This is a a creepmeter and across the hill we have the deep drilling site

And the reason Parkfield was the hub of earthquake activity in California

was because it had suffered from earthquakes with unusual regularity

The last one was in 1966

the prior one was in '34 and the one prior to that was in '22

but extending over a longer period of time they appear

to be roughly every 22 years

On that basis it meant that the next one was expected around 1988

By 1987 much of the equipment was in place and Parkfield was ready to capture

an earthquake as it happened

Then the wait began

1988 arrived.They waited and they waited,but the year passed without event

1989 nothing happened.1990 nothing happened.1991still nothing happened

Then in 1992 something did happen

California was hit by the most powerful earthquake in 40 years

The only trouble was it wasn't at Parkfield. It was more than 400km away

and struck the small town of Landers

The small desert community of Landers where the pavement ripped

and scattered like someone shuffled a deck of cards

Before today few people probably had ever heard of Landers,California.

Now the community has made seismic history

The largest amount of ground movement in a California

earthquake in over 100 years

Geologist Sue Hough felt the tremors from 100km away

The Landers quake where I felt it,it was like somebody picked up my house

all of a sudden and just started shaking it back and forth

It, the power in it is just remarkable

It comes out of nowhere and the,the energy involved is just staggering

One person died and 171 people were injured

There was million of dollars' worth of damage and the fact that Landers

was hit and not Parkfield seemed to show just

how unpredictable earthquakes could be

but Landers was not the dead-end it first appeared

In fact,it was to give scientists their best insight into how one earthquake

could set off another and it would bring them closer

to understanding the earthquake storm

because three hours after Landers a huge aftershock struck 40km away

near the town of Big Bear

Holy shit!

This footage was taken at a nearby observatory as the earthquake hit

What happened to the telescopes?

Destroyed. Go,go,go, come on.It may fall off.I got it on tape

Convinced the two events were linked,geophysicist Ross Stein

was determined to find the mechanism for how Landers had triggered Big Bear

The argument of people who held the earthquakes were independent and random

was basically to dismiss the presence of aftershocks

just say well let's just skim those off the record of earthquakes

and what's left over is random and that seemed ludicrous to me

because about a third of the earthquakes in any earthquake

catalogue are aftershocks

but when I began to look I saw that aftershocks were occurring all over the place

very,very far from the fault that ruptured and so aftershocks were the clue

that we needed to follow up,to figure out how earthquakes interacted

And his team had a great opportunity

Instruments around the state captured Landers and the aftershocks

that followed in extraordinary detail

The scientists knew how deep the rupture went

where the fault had broken and precisely how far it had slipped

They had numbers for everything.

Data is king in earthquake science.It trumps theory,it trumps everything

When we have great data we really can develop a deep understanding

Nothing exceeded or met the quality of data we had for Landers

Never before had their been so much information about two earthquakes

so they set about looking for any connection between Landers and Big Bear

They had been working on a computer model that they hoped

would illustrate how one earthquake could trigger another

The idea behind it was remarkably simple

Go,come on

As an earthquake strikes and the fault lines pull apart the system release

a huge amount of pent-up stress

Move it,come on

Their theory was that as the earthquake releases this stress it is redistributed

to areas close by seen here in red

These are the likely areas where the next earthquake should take place

the trick was to find out where these red zones might be

To do this they began to feed the data into their computer

The model had to take into account hundreds of calculations about

how much the fault had slipped,how deep underground the rupture had occurred

and the elasticity of the Earth's crust

So the way we set this software up was to give us a visual sense in which areas

were more likely to produce earthquakes and were more

hazardous were turning red and areas that were farther from failure were blue

Finally the model showed where the stress from Landers had been transferred

The area to the west lit up in red

We were hunting for the possibility that Big Bear occurred in a region

one of these red lobes where this stress was jacked up

by the occurrence of the Landers earthquake

And sure enough slap bang in the middle of the red zone sat

the Big Bear earthquake

Having Big Bear land in the middle of our red blob was very exciting

but that in itself doesn't clinch the case because Big Bear could have

landed there by accident and that's always a possibility and so what we were

to depend on were not just that one earthquake,regardless of its size

but the 3000 or 4000 other aftershocks that occurred in the first few weeks

Were those also occurring in the red zones?

Once again they returned to their model

They set about plotting every single aftershock that rippled out from the Landers

quake and,just like Big Bear,most fell in the red zone

We came away from this experience feeling that Landers had taken a possibility

a possible set of interactions and turned it into a probability

It now looked to us like this kind of earthquake interaction was likely

Landers was a scientific breakthrough

It gave the most dramatic evidence of how one big quake could trigger another

Landers was really a landmark event in giving us this incredible data set

which showed us how different large earthquakes interact and letting us explore

the mechanism of this interaction in detail

It was the best example so far of how the transfer of stress

could trigger a large aftershock

However,it did not show how one event could set off larger earthquakes across

longer periods of time.It did not prove the existence of an earthquake storm

To show this scientists had to test their model on somewhere

even more dangerous than the San Andreas fault

Turkey is in one of the major earthquake zones of the world

Here millions of people live under the threat of one of

the most treacherous faults on the planet

The North Anatolian fault slices through the top of the country for over 1000km

For centuries earthquakes along this fault have torn the region apart

In the last 100 years alone it has claimed the lives of nearly 100,000 people

Seismologist Celal Sengor knows it as well as any

The North Anatolian fault seems to display a cyclic behaviour

It looks as if earthquakes start at the east,they migrate west

then there's a period of quiescence.It starts again migrates west

there's another period of quiescence

This pattern heading from east to west attracted earthquake

hunters from around the world

Among them was Geoff King

He became fascinated by a study of the fault that suggested these earthquakes

might be triggering each other

This study began with an event that happened over 60 years ago

In 1939 a devastating earthquake struck the city of Erzincan in eastern Turkey

The earthquake that occurred in Erzincan in 1939 was the start of a sequence

of earthquakes that passed right along the North Anatolian fault

It was a major event that killed tens of thousands of people

Though there was very little data from the vent

there was enough to make an educated guess as to where the stress had gone

Using the same model that had shown the Big Bear and Landers connection

the red zone produced from Erzincan lit up an area to the west

Tragically three years later in 1942,an earthquake

struck near the town of Tokat.The model seemed to be working

After the Tokat earthquake stress was transferred to the west

The next earthquake was to the west

Then a series of earthquakes struck westward along the fault

the last in 1967 near Adapazari

The model showed that each event had occurred right

in the middle of the red zone

The earthquakes seemed to be triggering each other

like a set of falling dominoes

Following the Erzincan earthquake a series of events zipped

or rather unzipped the fault right the way along its length here

with stress being transmitted from one to the next

in a most spectacular fashion

These events were not aftershocks

and they were separated from each other by years

An earthquake storm appeared to be blowing across the region

but forecasting earthquakes after the event was relatively simple

A better test would be for scientists to use the model

to forecast where the storm would strike next

Our objective was to go from the 1967 earthquake to the west and to the south

and include all the other earthquakes for which we had historical

information in order to assess where stress had been transferred

and what was the likely earthquake future for those regions

And when he plugged in the data the red zone lit up over the Bay of Izmit

home to nearly half a million people

The red zone was right over Izmit and the Bay of Izmit and we knew

a major fault capable of catastrophic earthquakes

passed right through that region

Geoff and other scientists were so convinced the model

was working papers were published forecasting where

the next earthquake would likely strike

They could not say when,but they spelt out the specific threat to Izmit

A catastrophic earthquake seemed to us inevitable

and this had to be published

This was published in the scientific literature

but then it was also published in the Turkish popular press

the popular scientific journals, even in a newspaper

in the scientific supplement of a newspaper

I must confess even when I read that I said oh wonderful

you know,scientific prediction,but it did not occur to me to say my God

we've got to do something about it

The publication created only a ripple of interest

and for the people of Izmit life continued as normal

Most had no idea that the ground they now walked on

was under extraordinary stress

and then in August 1999 at just past three

in the morning Geoff King's forecast came true

45 seconds later much of the city and the surrounding area lay in ruins

The quake was so huge that buildings collapsed nearly 100km away

Despite the warnings the area was totally unprepared

Fire Chief Emin Pehlivan and his team were just not ready

I could say me and my team we did our best under the circumstances

and I am proud of that

We did a lot

We stretched resources to the limit

but this disaster caught Turkey off-guard

The rescue services had been given neither the equipment

nor the manpower to deal with the disaster

When we pulled out dead bodies or heard about them

it was equally devastating

There were times most of the time, we pulled out more corpses

than survivors and felt completely broken down

Each time we heard through the wireless that someone had been pulled out alive

a man,a woman aged this or that,we felt wonderful

All of us in the headquarters shouted out of joy

as if we had won a victory and hugged each other

Among those trapped was ?m?r Kinay

She and her mother had been caught in one of the many collapsed buildings

I tried to make out the voices I knew.I heard babies crying

I was lying on top of my mother.I was trying to hold her hand

talk to her but she didn't respond

This is the very moment of her rescue

She was finally found nearly three hours after the fault had ruptured

but her ordeal was far from over

I warned them when they were digging close to my head

Then they dug the area below

I could move my left hand I managed to get it out through a hole

and they dug around it

They removed the concrete with a crane

removed my mother

My hair was still under the rubble The man started crying

He said we can't get you out my child,your hair is trapped

I said cut it He kept crying and said I can't cut this beautiful hair

?m?r was rushed to hospital knowing nothing of her mother's fate

Her visiting family kept any news from her

I learned about the loss of my mother after about 40 days.

She was the one looking after me

One day I said,I asked her to take my Mum's photo

They didn't even bring me a photo She started crying

My cousin left the room and I realised Mum was dead

The exact number of casualties from the Izmit earthquake is still not known

but some estimates suggest that perhaps 25,000 people died

Hundreds of bodies are still unaccounted for

but for Geoff King, one of the scientists who had made the forecast

the Izmit earthquake was a bittersweet moment

He had been proved correct

but he now realised that use of the model came with enormous responsibility

As the scale of the disaster became clear you know

we were clearly right in our prediction, but this is on the one hand do

you want as a scientist to be right

but on the other hand you hope you're not right because you realise

that the red areas that you calculate represent many deaths

tens of thousands of potential deaths

Now it really did seem that an earthquake storm was heading across Turkey

. It was vital to forecast where it would hit next

Well after the Izmit earthquake we had important new data

I mean we could immediately expand the studies that we were engaged in to see

what would happen next and what was likely to happen next

so following the earthquake we immediately accelerated our work in this area.

Geoff King and his team set to work to try and find where the stress had moved

As they put the new data from the Izmit quake into the computer

an area to the west again beamed red

This time though things were not so straightforward

Izmit had been forecast because scientists knew the exact nature of the fault

running through the area,but now they could not tell

The fault that moved in the Izmit disaster passing under this road.

It then at this point it goes off-shore under

the Marmara Sea and completely disappears

The extent of any danger in this newly stressed red zone

depended on understanding the exact nature of the fault

Before the 1999 Izmit earthquake happened

we knew very little about the Sea of Mamara

The North Anatolian fault we know that enters at one end exists the other

but in-between we really didn't know much more

What made this all the more urgent was the fact that

when they ran the computer model at the edge of their newly plotted

red zone was one of the greatest cities in the world:

Istanbul

With so much at stake it was crucial to understand the fault in intimate detail

Just how big a threat was there to the city?

Soon the Italians,French,Turks and Americans

set about making the fault at the bottom of the Marmara Sea

one of the best known on the planet

The first thing we thought we had to do was to map the sea

to know where the different features on the bottom because

in a sea the ocean is not active like on land

so you preserve the trace of faulting in a much better way

Gradually the fault began to reveal its secrets

Mud cores taken from the sea-floor revealed the huge quakes

that had struck the area in the past

and the seismic signals indicated that

the whole area was still very much alive

Other results showed just how vast the faulting systems was

We,we found this long valley that cuts through the whole two-third

the western two-third of the Mamara Sea a spectacular valley

that we did not expect and that was obviously a very active fault

Although a picture of the fault's activity was emerging

the exact structure of the fault remains unclear

but what was known was that an active faulting systems stretching nearly

the length of the Mamara Sea sat a mere 20 kilometres from Istanbul

A big rupture along this fault whatever its precise nature

could be a massive disaster

The earthquake that we expect would happen close to Istanbul will be as big

or bigger,than Izmit. Izmit killed 20,000 or even 30,000 people

The population density around Iz,Istanbul is ten times greater

The scale of the catastrophe to a European city is almost unimaginable

What makes this forecast so worrying is that many buildings

have not been built to withstand a disaster of this kind.

We are now in one of the poorest parts of Istanbul

You can imagine that the buildings around us are not of the best quality

When the earthquake hits parts of these buildings will fall down

people will be trapped in them.Some of the people will be able to get out

Some of these streets have natural gas pipelines going under them

They will burst right. You'll have fires rising

People will try to run away,they'll be ruined,they will try to clamber up

Rescue units will try to reach them.It will be complete

mayhem in Istanbul if that happened

No one can yet tell when the earthquake storm will strike

The forecast simply cannot do this.It could be in 100 years time

it could be tomorrow

The scientists have at least identified where it is likely to hit next

This means there is still time to prepare

The stress has been building up in Istanbul

We know there's going to be an earthquake

We don't really know when there's going to be an earthquake

but we know it'll be a major earthquake

Buildings can be improved, construction can be modified

emergency services can become better organised

There are very many things that can be done and this will bring

the death toll down by ten times, or even 100 times

and it is completely possible and it is economically feasible

So the forecast comes with hope

There is a chance this time that the people of Turkey

can be prepared where the disaster strikes