The earthquake happened at midday, and people out of doors had difficulty in standing up.
It was the strongest tremor ever recorded in northwestern Japan, an area of farming and fishing villages.
The epicenter was a hundred miles out to sea, but the whole northern part of Japan was shaken.
Smaller shocks have continued throughout the day.
Each year there are several hundred major earthquakes in the world.
Are we any nearer to predicting them and reducing the havoc they cause?
And indeed, might they even be of use to us?
Late one afternoon in the early summer of this year, geophysicist Rob Cochran was driving home from his work at the United States Geological Survey in California.
The bleeper was triggered by a computer, the same computer that generated this picture.
Connected to hundreds of seismographs all over California, it processes information about earthquakes.
Here it's building up a picture of the swarm of earthquakes that had hit the small town of Kowalinga in central California.
Activities all around, right around there.
If the shocks had looked as though they were the forerunners to a very large earthquake, the Geological Survey would have notified Washington, who would have had to decide whether to evacuate.
There's the latest one.
The largest is about a magnitude, something larger than a five, probably close to a six.
Things are probably going not too well down there.
This, though, was only a moderate earthquake.
Although there was a fair amount of damage, nobody was killed.
Distressing, though this is, to those who have to endure it, earthquakes can reveal a good deal to geophysicists about the earth we live on.
And some of this information, as we'll see, is of use to countries like Britain, which is no longer an earthquake zone.
And while scientists struggle to understand and one day predict earthquakes, it's also becoming possible to reduce the vulnerability of people who have to live in areas of the world prone to them.
In Kowalinga, most of the damage was to old buildings, put up before modern construction codes were introduced.
It's the same in parts of Chinatown in San Francisco.
Contrary to popular opinion, it's these old low-rise structures which are far more vulnerable than the high-rise close by.
At the beginning of the century, the technique of building skyscrapers had only just been introduced.
But during the 1906 San Francisco earthquake, the few that had been built withstood the shocks well.
Although caught in the subsequent fire, the famous San Francisco Hotel survived and is still in use today.
The secret of these buildings' strength is in the type of construction, a framework of steel girders bolted together.
A mathematical model of what would happen to such a building during an earthquake exaggerates the horizontal movement, but shows that while it would sway alarmingly, the building would survive.
During an earthquake, the greatest danger with such buildings is when the glass from hundreds of shattered windows rains down on the people below.
And the greatest risk of all is during the rush hour.
Aware of the problem, the National Guard regularly rehearse rescue procedures during earthquake alert exercises.
In San Francisco, it's estimated that up to 11,000 people would die, and in Los Angeles, 23,000 people.
It might seem a lot, but it's nothing compared to the effects of a similar-sized earthquake in the Middle East or Asia.
Here, hundreds of thousands of people might die.
The rescue resources available make a difference, but the initial death rate depends on something else, the design of the buildings.
The Civil Engineering Department of Imperial College, London.
Telltales, small pieces of glass, are being glued to a section of wall.
The wall is mounted on a shaking table. It's a means of reproducing an earthquake in the laboratory.
The source of it will be a couple of hydraulic rams.
What they're going to do is subject the wall to a modified earthquake.
In this case, the wall will have to endure the El Centro, California earthquake of 1941.
A computer stores information about the earthquake and translates this into commands to the rams.
Vertical movements broke the bonding between the bricks and mortar, and the lateral forces eventually toppled the wall.
The engineer in charge of the project, Professor Ambraces.
One of the things we are trying to do here is to see how you can improve brickwork and brick construction in order to be able to withstand shaking after a certain degree.
Brick is the building material which is used by more than 80% of the population of the Earth.
I mean, you can't really ban brick in seismic areas.
You have to use brick, but you have to use it in such a way that you can withstand large levels of shaking.
The idea is to limit the breaking of the bonds between the bricks and mortar.
The solutions are not new.
There's evidence that thousands of years ago, the Cretans built the palace at Canossos with wooden horizontal beams to contain movement during an earthquake.
For the last few hundred years, this has been the traditional type of building in parts of Turkey, close to the Anatolian Fault.
It's made of bricks, but the first step is to build a wooden framework into which the bricks are laid.
The modern equivalent is to build in beams of reinforced concrete either vertically or horizontally every few courses.
A cheaper version might involve merely laying chicken mesh in the mortar between the bricks.
At the end, I mean, you are designing for the average earthquake that's likely to happen during the lifetime of the building, which has a limited lifespan.
And for the rare one, which has a chance once in a thousand to happen, perhaps the building will be completely shattered, but will not collapse and will not kill the people.
But there is another way of reducing casualties, and that's to predict the earthquake before it happens.
Our knowledge of why and where earthquakes happen is linked to a theory devised about 15 to 20 years ago, which explained the movement of the continents around the world.
The theory of plate tectonics showed that the world's surface was made up of huge, rigid plates of rock.
These plates slowly moved around the world, carrying the continents with them.
At the boundaries where these plates are in contact with each other, they produce earthquakes.
If the earthquakes of the world are plotted, they effectively mark the plate boundaries.
This thin red line of dots, for example, shows where earthquakes have occurred in the Atlantic.
They mark precisely the boundary between the American and the European and African plates.
Because plate tectonics can reveal how fast and in what direction those plates are moving, it's potentially of use in predicting how frequently earthquakes will occur.
Take the so-called Ring of Fire around the Pacific, a string of earthquakes and volcanic activity at the plate boundaries.
As far as the Californian part of the Ring of Fire goes, it's known from plate tectonics that the Pacific plate is moving relative to the North American plate at five or six centimetres a year.
But it's only a gross average deduced from millions of years of movement.
It's generally thought that the next very large earthquake in California will be here in Palmdale in the high desert.
But the problem with plate tectonic theory is that it's not precise enough to say when.
To get a better idea, down on the fault itself, they've cut through the bed of an old peat bog to reveal a history of 1,500 years of earthquakes in the broken layers of peat.
By dating them, it's possible to arrive at a more precise average, and they've found that large earthquakes occur about once every 150 years or so, give or take a few decades.
The last time was 1857, so the large city just over the hill, Los Angeles, could get hit next year or in another half century.
It's a prediction of a kind, but not very precise.
At the United States Geological Survey, they hope to make prediction more exact by recording all the seismic activity that looks relevant, such as the buildup of a pattern of smaller earthquakes.
They then hope they can deduce what it all means, but that's not as easy as it might sound.
Take what's happening here at Mammoth Lakes, where the well-off of Los Angeles flock for the odd skiing weekend.
High up in the Sierra Nevada mountains of California, it's undeniably beautiful. Yet this beauty is to some degree the result of a cataclysmic event that occurred three-quarters of a million years ago.
An enormous volcanic eruption produced the huge crater valley, stretching 20 miles across from the snowy mountains in the distance to the rim of hills in the foreground.
What caused it?
According to plate tectonic theory, a few tens of million years ago, the floor of the Pacific Ocean was spreading out and diving underneath the North American plate.
In doing so, the movement produced hot fluid rock under the western edge of North America.
Some of this escaped to the surface, producing volcanic activity.
Although these movements stopped about 17 million years ago, that hot rock is still down there and could burst through again.
It could happen here at Mammoth Lakes.
In the center of the valley, formed by that 700,000-year-old eruption, the land has risen in a small bulge, caused, it's believed, by upward pressure of the hot fluid rock.
The problem is to predict if and when it will break through.
The signs that hot rock is close to the surface of the earth are all around in the valley.
Hot springs are everywhere, and seemingly innocent pools are, in fact, scaldingly hot.
Because of this spectacular and beautiful scenery, the tourist industry thrives.
The area of Mammoth Lakes has become a modern-day western boomtown where the wealthy come to buy their homes.
Or rather, they did until back in 1978, when something happened to deter them. Earthquakes.
It happened again in 1980. By Californian standards, they were not big. They merely emptied the bookshelves.
But they were sufficient to attract the geological survey into the area.
The reason being that these earthquakes might have been signaling another massive volcanic eruption, which could wipe out the town.
The scientists buried seismographs and tiltmeters in several locations throughout the valley in an attempt to get a better picture of what was happening.
With the aid of solar batteries, information from these instruments was radioed back to the United States Geological Survey, though just what it all meant, nobody was sure.
Here, the computer replays the earthquake swarm that the instruments detected in January of this year.
At the time, it created some concern amongst the local people, but there have been several such swarms in the last five years,
and in between, as now, everything appears to stabilize.
The question is whether this pattern of stop-go-stop-go means that something very big is about to happen down there.
In an attempt to find out, geological survey scientists are applying other techniques to the problem.
In the gathering dusk, they've set up, right on top of the bulge in the center of the crater, a sophisticated laser ranging device.
On a far hillside, a colleague holds a hand torch.
They'll use it to line up the beam.
As the sun goes down, it becomes dark enough to use the low-power laser.
On the far hillside, a mirror has been mounted on a tripod, and a colleague gives directions by radio.
The idea is to shine the laser beam onto the mirror, so that it's reflected straight back to the transmitter.
By measuring the time it takes for the beam to return, it's possible to work out exactly how far away the mirror is.
When the distance is measured on successive nights over a period of time, any movement of the land itself will be picked up.
Two lasers, red and blue, are used simultaneously to correct for problems caused by pressure and temperature changes, which could affect the precise measurements.
What these and other techniques show is that the ground is indeed moving.
The western end of the valley has risen 40 centimeters in three years.
But once again, what it all means, no one is quite sure, except that it is potentially dangerous.
One of the chief scientists working on the project, Mark Zoback.
There are 35,000 people essentially living on a volcano that's active in a geologic sense, if not in a historical sense.
So the health and safety of those people are our foremost concern.
And while we tried to be very conservative in our estimates and not make any irresponsible statements to cause unnecessary alarm,
I think the fact that the seismicity is going on is a cause for concern.
We've certainly stepped up our efforts in that area, and the public knows this.
They certainly do, and it concerns them.
Estate agents, for example, are having a bad time.
Since the swarm of earthquakes last January, they've seen a drop in business activity.
Real estate agent Chuck DiMarco.
We had, oh gosh, more activity in the month of December than we'd had in, say, the previous five months.
But about 80% of those people decided that because of the possibility of maybe a major eruption or a large earthquake that they decided not to buy.
Others have decided to close down their businesses. Kerry Walters.
I don't know the total, but I know that at least 56 businesses in this community have been foreclosed in the past four to six months.
Now this compares with, I've lived in Mammoth Lakes for seven years, and I know that in at least the last five years,
to the best of my knowledge, only three businesses have had to go through foreclosure.
So this is quite a giant increase.
As far as the skiers are concerned, it doesn't appear to have bothered them very much.
They are, after all, risk takers.
Nevertheless, for the people who live in the valley, prediction is a mixed blessing.
They feel that the scientists should continue their work, but they want something more precise.
It's the uncertainty of it all that's a drawback.
Well, there may be drawbacks to prediction, but they are far offset by the benefits.
If the situation at Mammoth Lakes had gotten considerably worse instead of having stabilized as they did,
it would have been a very good thing that as many people were leaving as fast as they could.
So what we try to do is give to the public the most accurate information about what's happening that we can,
and try to interpret that data as accurately as we can.
And that's really all we can do.
I don't think they know what they're talking about.
I don't really view it as a strict science.
I think it's a neophyte science that has got a way to go.
In its infancy, they're going to make a lot of mistakes, and this has probably been one of them.
The judgment is too harsh.
The problem is that plate tectonics gives an idea of what's happening globally,
but it's not precise enough to explain exactly what's happening locally here.
The scientists know in principle what's happening deep down,
but not how it relates to what's happening at the surface,
and prediction depends on being able to do just that.
But our theories of how the land moves are being refined,
and on the other side of the world there's the beginnings of a more precise understanding of what's going on.
For millions of years, Africa has been moving steadily north and colliding into Europe.
The Mediterranean will probably disappear one day and be replaced by a huge mountain range.
So far, it's formed the Alps.
The Mediterranean has already dried up once before.
About five or six million years ago, it became closed at both ends,
and the sea evaporated, leaving a salt-encrusted basin.
Later, the fresh water running into it gave rise to oases with plants and animals.
Quite how it all ended isn't known.
It's possible that earthquakes broke open the Straits of Gibraltar, and the Atlantic poured in again.
What is certain, though, is that according to plate tectonic theory,
Africa is still moving north into Europe,
and this would imply that all the land along the whole length of the Mediterranean is being compressed.
But that is not what is happening in Greece, and once more the clue came from earthquakes.
Athens on the night of February 24, 1981.
In the old observatory, the seismic activity from all over Greece is constantly monitored.
It was 10 to 11 local time, and the only movement was from a nearby seismometer
picking up vibrations of the Athens traffic.
The alarm had been triggered by an earthquake near Corinth.
The trace had apparently stopped.
In fact, the signal was so strong that the pen had swept off the drum.
A television crew driving out from Athens picked up the first signs of the earthquake about halfway to Corinth.
The chief seismologist at Athens Observatory, Professor Drakopoulos.
Many buildings in Greece are designed to withstand earthquakes,
but inevitably there were a few that failed.
Once again, these tended to be the older type of building.
Katerina Iannou lives in a nearby village hit by the earthquake.
I was asleep when I heard the noise.
My house moved like this.
Holy mother, holy mother.
I went to the door, but instead of unlocking it, I was locking it.
At last I got it open, and I heard the neighbors calling my name.
Katerina, Katerina, there's been an earthquake.
You must come away from your house.
Come to the square with us.
In the square, the priest told us all to sit down.
They took the seats out of Sotiris Cafe, and we sat, but it was raining, so they lit big fires.
The priest was making tea and making coffee, making all sorts of things to give to us.
All through the night the earth was trembling gently, sweetly.
And then at four in the morning it happened.
The earth moved angrily, and the priest said,
On your knees, everyone on your knees, and pray.
And we knelt and prayed, and we cried, and slowly it stopped,
and at daybreak we started to move.
Earthquakes have been happening in Greece for a very long time.
The scenery itself has been formed by them.
During the Corinth earthquake, some of the mountains around here rose upwards by three or four feet.
But these very earthquakes, combined with the fact that the area is so seismically active,
and the faults are so easy to see,
made it possible to get a much clearer idea of how what's happening at the surface
relates to what's happening deep down.
The first thing to do is to examine the faults themselves.
One of a group of people working on the Corinth earthquakes is Dr James Jackson, a Cambridge geophysicist.
This is a fault which moved during the earthquake.
The fault, which resembles a knife cut, goes down about ten miles into the earth.
What we're looking at here is the mountain, this rock,
which has been punched up out of the valley floor during the earthquake.
In fact, the valley floor that I'm standing on now was here before the earthquake,
and you can see where it was by the red stain on the rock,
which comes from the soil in this valley, which is red.
It had done it before.
Up here, this was also below the valley floor at one time,
and has been uplifted in previous earthquakes a few hundred years ago.
Now, to give some idea of the size of this fault,
it goes down, I've already said, to ten miles.
In this direction, it goes along about ten miles.
Now, this was not even a particularly large fault.
The one which moved in the San Francisco earthquake was about a thousand times larger.
And yet, when you think about the energy needed to create this mountain,
or to start to build this mountain, this is a young one,
that energy is enormous.
The energy released in this particular earthquake was larger
than the largest nuclear explosions ever detonated by man.
These villages around here were right next door to this fault here,
which is equivalent to about a few million tonnes of TNT being exploded next to them.
So it was clearly very terrifying,
and their descriptions of what it was like really bear this out.
They were thrown to the ground.
They couldn't get up.
They were just bounced around for a few seconds, perhaps a few tens of seconds.
A clue to what's causing these earthquakes is that when these faults move,
they do so according to a pattern, and it's easier to see that down at the coast.
The interesting point here is what's going on down at the water level.
In this particular bay, all the rocks are overhanging.
The formation is known as a solution notch.
The reason this cliff here projects out in an overhang
is because the lower part of it has been dissolved away by the seawater.
This is happening now, over there, below sea level.
This has not been caused by wave action.
It's happening below water very slowly.
In ten years, you would notice nothing.
In a hundred years, perhaps a little bit.
This has probably happened over a thousand years.
If this has happened below water,
you may ask, what is this doing up here now?
And not only this one, you can see higher up.
There's another overhang coming down like this.
The reason these are now out of the water is that this whole cliff has been uplifted
here at least eight feet by successive earthquakes,
probably over a hundred or few thousands of years.
You can trace these overhangs at roughly the same height,
right round this part of the coast.
We can get some idea of how long these rocks have been lifted out of the water for
by taking a look at these holes here on the side of the rock face.
These holes were made by mollusks, a sort of shellfish,
drilling into the rock when it was below sea level
because that's where these mollusks live.
They don't climb out of the water to do this.
Since then, obviously, the rock has been lifted out.
Now, inside these holes, there are still fragments of the mollusks remaining.
The animal has long since been dead,
and what we see are fragments, the remains of dead shells at the front of the holes.
And we can take pieces of those dead mollusks
and date them using radiocarbon techniques.
This has been done by people at Cambridge and London,
and we now know that this bit of rock was lifted out of the water less than 10,000 years ago.
It's obviously been going on much longer than that
because we can see higher up by that solution notch older holes
which were lifted out of the water long before these ones.
So this has been going on for several thousands of years at least.
So the earthquakes are lifting the rocks out of the water,
but only a few miles along the coast, the opposite seems to be happening.
The ground here is sinking and becoming marshy.
This small fishing village used to have a beach,
but since the last earthquake it's sunk, as the whole village will be one day.
And there are no solution notches above the water level.
Here the earthquakes are making the land sink.
The phenomenon has been known for a long time, but not in quite the same way.
From the air it's possible to see in the shallow water along the coast
the faint outlines of streets, houses and temples.
Sunken cities lie all around the Mediterranean,
but they're particularly prevalent in the east, especially around the Aegean.
Some of them date back two, three or even four thousand years to Neolithic times.
Until recently, one of the most popular explanations of this has been a change in sea level.
Not far from Corinth is the ancient port of Kekrie,
a network of long-submerged streets and buildings lies just under water.
Down at the water level itself, only the remains of a few foundations can be seen.
We're standing here at Kekrie, which is an ancient site where, about two thousand years ago,
St Paul preached a sermon in a church just over here.
Since that time, the whole area appears to have sunk by about two meters.
People used to think that this apparent rise in sea level
was caused by melting of ice at the end of the last ice age.
However, just round the coast a few miles are solution notches.
And to explain these solution notches in terms of sea level changes,
one would have to accept that the sea level had fallen.
It's obviously impossible to have the sea both going up and down in virtually the same place.
The only explanation is that the land moved, not the sea.
What's more, it moved in a regular pattern right across the Aegean.
Why was a mystery, until in the last few years Jackson and his colleagues came up with an explanation.
This new theory suggests that millions of years ago,
the land now called Greece was stretched out into what is now called the Mediterranean.
As it got thinner, it sank down and as a result of this, it flooded.
So rather than a solid piece of land, the stretching produced a shallow flooded basin,
now called the Aegean Sea, dotted with the Greek islands.
It's these islands that provide a clue to the way in which the land stretched
and the reason why there are earthquakes.
If you want to stretch the top of the earth's crust,
you can't do it like stretching a rubber band or a piece of plasticine or chewing gum.
It's too hard and it will break.
In fact, it breaks along faults and that's what earthquakes are.
When these faults slip, you get an earthquake.
Let me show you this block of wood floating in a fish tank.
In fact, this block of wood represents the earth's crust,
which is floating on fluid hot rocks beneath.
Everyone knows hot rocks are orange.
Now, let's have a look at this more closely.
This block of wood represents a piece of the earth's crust in Greece.
What I'm going to do is stretch it by giving it an earthquake
and I'm going to give it an earthquake by moving it along a fault,
which I've conveniently sawed through the Greek countryside.
Before I do that, look at this black line here.
It's continuous across the whole piece of wood.
Now, during the earthquake, what happens is the fault moves like this.
So the side has moved up, the side has moved down.
You'll see that the black line is now no longer continuous
and that the ends of the black line are separated by about an inch
in the horizontal direction.
So the block of wood has got longer by about an inch.
It's stretched.
Now, in Greece, there's one of these large faults every 20 kilometers or so,
so this block of wood would have several faults right across.
Now, each of these faults moves probably a few kilometers
and is responsible for a kilometer or so of extension.
So right across the width of the Aegean,
you have a tremendous amount of stretching.
The question is, what will this configuration look like
when it floats on the fluid rocks beneath it?
Well, here are fluid rocks in the orange fish tank.
Let's find out.
Well, now we've put the crust back in the fluid rocks beneath
and we've made ourselves a Greek island.
Here it is.
You can see what's happened by looking at the black line,
which was level with the top of the fluid before.
This side, the Greek island, has moved up.
In front of it here, we have an escarpment,
which is the fault escarpment.
And in front of the fault escarpment,
we have this block, which has moved down,
and that is the deep trough,
which is offshore in most of the Greek islands.
The crust would be much thicker than is shown in this diagram,
but the principle is the same.
It is formed when the sea floods in.
All this is due to stretching.
The islands that James Jackson is talking about
lie in the northern and central areas of the Aegean.
The island of Alonissos is one of these,
and it matches well.
A gentle slope to the right and a steep escarpment to the left
with a deep trough just offshore.
And the process is still going on.
It's exactly what the earthquakes
at the sunken city of Kekrie, near Corinth, are doing.
The position of Kekrie is very similar
to the deep water troughs next to the islands of the Aegean.
Where we are standing now is still partly above sea level.
But if this fault, which is next to the big mountain over there,
continues to move, where we are now will continue to sink,
and after some time, maybe a million years,
will be a long way below sea level.
Whereas that mountain there will be rising
and eventually will be an island.
So, despite the fact that plate tectonics is right
in telling us that Africa is crunching northwards into Europe
and the whole area is being compressed,
the Aegean, at least, is an area that is being stretched.
It's a huge basin.
This explanation of why the land goes up and down
gives us a clearer understanding of how earthquakes happen here.
But for earthquake prediction, it's still not gone far enough.
Now, these observations of the coastline going up and down
are very helpful in guiding us to where precisely the large faults are.
And to that extent, these observations are useful.
But what they won't do is tell us when the large faults are going to move.
In order to predict when a fault is going to move,
we need to know what to look for before faults move.
At the moment, we can't really do that,
and I don't think we will be able to do it
until we know much better what goes on during
and especially before earthquakes.
Progress in that direction has been slow, to be honest,
and my personal opinion is that we won't be able to make
useful predictions before the end of the century.
So, for the present, prediction of earthquakes seems a long way off.
But there is something else.
This work is especially important to us in Britain,
and that's because all this stretching and island formation
has happened before much nearer home.
The North Sea.
A hundred or so million years ago,
the North Sea and indeed the whole Atlantic
looked very different from the way they do today.
About 150 million years ago, the South Atlantic had started to open,
but the North Atlantic didn't exist.
Greenland would have been a lot closer to the British Isles,
Newfoundland was quite close to Western Ireland,
and in particular, British Isles were much closer to Scandinavia
than they are now.
Well, soon after this time, America started to move westwards,
away from Europe.
Now, as that happened, it stretched the whole of this area,
rather like this.
And eventually, this area broke,
but it didn't break along the North Sea here,
it broke west of Ireland.
Having broken, it moved westwards rather like this,
taking Greenland with it.
Although the map exaggerates everything,
the important thing is that the British Isles moved westwards,
away from Scandinavia.
If you remember, they started off further east, like that,
and during the stretching, the British Isles moved westwards.
Now, that process, that moving westwards,
is what is responsible for the North Sea.
The North Sea will have stretched, it will have become thinner,
and it sank.
According to James Jackson and other Cambridge scientists,
the North Sea was stretched in the same way the Aegean has been.
The land of those days, marked in the lighter colour,
includes islands formed in the same way as the Greek islands.
But as well as North Sea islands, something else was formed.
Oil.
America had failed to take the British Isles with it across the Atlantic,
but left this as a parting gift.
The reason is quite simple.
The North Sea became a sedimentary basin.
Because of America's lingering goodbye, the crustal rocks,
which were destined to become the North Sea, stretched.
As they did so, they thinned, and just like the Aegean, they sank and flooded.
This warm, shallow sea became an attractive home for bacteria
and simple plants that lived and died in their billions.
The sediments of corpses became squashed and compressed into rock.
The hot rocks below were now closer to the surface.
Their heat permeated upwards and cooked those dead bugs
at about the temperature of a cup of tea.
The bugs contained hydrogen and carbon atoms,
and the cooking or maturation process slowly changed these hydrocarbons
into the type we call oil.
Some of this escaped upwards into the sea and was lost,
but much was trapped.
One reason was those islands that used to exist in the North Sea.
They were formed by fault movements just like the Greek islands.
Although they've long since sunk, that fault movement occasionally
positioned rock, which was impermeable to oil,
in such an alignment that as the oil rose to escape,
it could get no further.
Millions of years later, find one of those traps under the bed
of the North Sea, and you may have found an oil field.
But just how far did the North Sea stretch?
Some recent ideas of the Cambridge scientists suggest
that it's a lot more than had hitherto been thought,
and if true, this is important for the oil companies.
One of their consultants is professor of geophysics
at the University of Texas, John Slater.
I think the reason why the oil companies are interested in this type of research
is because it's going to affect their ability in the future to find more oil.
The reason why this is the case is because the amount of extension
is going to affect two critical parameters in terms of finding oil.
They're not the only parameters, but they're two very important ones.
The first one is that if you stretch a basin by a large amount,
you're actually going to increase the amount of heat that's put into that basin.
This is going to cook the oil early in its history,
and it's going to change the places where you're actually going to find the oil eventually.
The second reason why it's important is because if there's this amount of extension
that is being claimed by the Cambridge group, then what's going to happen in this case
is it's going to change the way the faulting goes in the basin,
and it's in these faults and the traps that are associated with these faults
that much oil is actually found.
But what led the Cambridge scientists to question how far the North Sea stretched?
One piece of work involved positioning underwater seismometers on the bed of the North Sea
whilst explosive charges were detonated.
The seismometer contained a cassette recording of the echoes of those explosions
as they reverberated around the floor of the North Sea.
Based on the results of the work, a computer draws a picture of the sediments of the North Sea
and the bottom of the stretched crust.
The paths of some of the shock waves went down through the sediments to the top of the crustal rocks
and then up to the seismometers.
Others went deeper down to the bottom of the stretched crustal rocks and then up to the recording instruments.
By measuring the delay and strength of these echoes,
the scientists were able to compute the thickness of the North Sea rocks.
What they found was that the crust was about half as thick as it would have been
had it not been stretched at all.
This result was one piece of evidence that suggested a major extension had taken place.
Further evidence for how far it had stretched came from the oil rigs themselves.
In order to keep the drilling bit lubricated, a slurry is pumped down to flush upwards
all the chippings of rock being chewed up by the bit.
And these chippings carry important information about what was happening to the bed of the North Sea
millions of years ago.
The slurry is filtered and samples taken to the laboratory at Cambridge.
The slurry is washed, filtered and dried.
With the aid of a microscope, it's possible to see the remains of small sea creatures
that died millions of years ago.
Different types of animals lived at different depths.
By studying them, the scientists can tell how deep the water was at a particular time.
And by taking successive samples, it's possible to learn how fast the North Sea was sinking.
The faster and deeper it sank, the more it must have been stretched.
Dr Rosemary Wood.
What we were able to see was that the amount of stretching varied across the North Sea.
On the edges we were seeing an extension of maybe 10 or 20 percent,
but towards the middle of the North Sea we were seeing much higher extensions of maybe 35, 50 percent.
So between say 180 and 120 million years ago, England moved away from Norway by about 50 to 80 kilometres.
This is twice as much as the oil companies estimate.
They work largely from seismic reflection data, which reveal sediments and faults under the seabed.
And by measuring the extension on those faults, they arrive at a figure for the total stretching across the North Sea.
But this data has been criticised by the Cambridge group as being inadequate,
because the measurements do not go deep enough.
Who's right? Professor Slater.
It's my feeling, and I think quite strongly, that in fact the Cambridge scientists are correct,
and that there's evidence not only from the work that they've been doing in the North Sea,
but work in other basins which are of a similar structure of style,
that what their approach and the amount of extension that they're suggesting is probably more likely to be correct
than the kind of restricted numbers that have in the past been suggested by the oil industry.
Now what has happened is that in fact not just the company that I've been working with,
but other companies are starting to take these ideas very seriously,
and they're starting to question their interpretation of the data.
And I think that the ideas that have been pushed by the Cambridge group and by some other groups as well
could have a major effect on the kind of ideas or concepts of how much oil is actually in places like the North Sea
or other basins in the world.
How does all this affect the oil reserves in the North Sea?
If it's true that the currently accepted ideas about the stretching are wrong and the Cambridge scientists are right,
it might mean that there's more oil available.
If so, how much more and what's it worth?
We don't really know the answer to that question.
However, let's take some speculations and say that maybe we find as a result of this work,
say four or five medium-sized fields,
then you're looking at an increase of about 10% in the oil in the North Sea,
and you're looking at essentially something like 2 billion barrels.
If you look at $35 a barrel, you're looking at $70 billion.
There are people who would say that, in fact,
most of the companies in the North Sea probably have had the wrong type of model
and that they could as a result of this thing find 10 or 12 medium-sized fields.
If that's the case, I would expect that the reserves would go up quite substantially
and you might be looking at a number in the order of 20 or 25%.
When you're looking at numbers like that,
the numbers that you're talking up in terms of dollars are $150 to $200 billion.
And even a small field is a very substantial economic find.
Although our increasingly refined theories of how the Earth moves
can't yet accurately predict earthquakes,
it's an intriguing thought that ideas that were developed to get a clearer understanding
about what's happening in the Aegean with its earthquakes
may have turned out to be useful in the North Sea.
And according to John Slater, these new theories may one day go even further than that.
Where this is going to be an advantage is not just in a place like the North Sea
where we can look over a large quantity of data.
There's going to be much more of an advantage in an area,
a frontier area where we're out in exploration,
like say off the coast of the Shetlands
or say in the area where there's just been a large major find
which is off the coast of Newfoundland.
I think the most important portion of the work that's being done now by the Cambridge Group
and others like them is that I think we're going to get out of this
a really good quantitative model of how these basins are formed.
And if that is the case, then it's going to have a major effect on the industry.
Thank you.