Inside Weather Science: How We Predict and Prevent Global Disasters | FULL DOCUMENTARY

Science and life
122 followers
The year is punctuated by weather events. Winter brings its storms and with them the challenge
to meteorologists to predict which coastlines will be hit and what the risk will be to local
populations. The lives, safety, well-being and economic activity of millions depend on their
forecasting skills. Today meteorological monitoring and forecasting are critical
for our very existence. But how do we predict the weather? For a long time the mechanisms
determining the weather were a mystery, until that is the invention of the satellite.
The launch during the 1990s of the two first satellites in the METOP constellation
revolutionized this science, allowing us to observe our planet from space.
Before we only had geostationary satellites at an altitude of 36,000 kilometers which only
show the equatorial regions. With METOP we can travel around the earth and see it all,
including the poles. We are at 840 kilometers so we're much lower and can take pictures and
much more precise measurements. For its third launch the METOP satellite was transported to
to be blasted into space on a Soyuz rocket.
And everyone involved in its design and production met up in Darmstadt, Germany to watch.
Space is never easy and a launch is never relaxing.
Anything can happen, good and bad, and we have to be ready for anything.
Satellite launches are rare events. In the Yumitsat control room the tension and anxiety
among the technicians who will soon be operating METOP is palpable.
I always bear in mind on a night like this our quote from Alan Shepard,
who was the first American into space, and his quote was,
Dear God, please don't let me fuck up. Don't fuck up.
It's wonderful to see all the technology behind a satellite launch and in the end we benefit
because we receive all the data. It just happens. We don't need to think about it.
But when you see all the technology behind it, it's absolutely fantastic.
Today 80% of the data used by weather forecasters comes from satellites.
But these instruments, which only see a very small part of the earth as they move,
are still too few and far between.
There are only 16 weather satellites in orbit around the planet.
And there are also around 60 research satellites that supply information to forecasting centers.
Weather forecasting has become a global challenge requiring huge resources.
And climate change is only adding to the difficulties.
Scientists are continuing their quest for understanding across the globe.
When the ball drops in Times Square, the wind chill could be minus four.
I have never felt the cold like I'm feeling it in these last few days.
For the past few years,
waves of cold weather have been regularly hitting the US, Europe and Japan.
But what is the reason for these events?
According to some scientists, arctic warming could be the culprit for this shift.
Unfortunately, this geographical zone is difficult to access,
and it is rare to obtain meteorological data from there.
Meteorologists have understood that to improve their forecasting,
they need to better represent this region in their modeling.
This is the start of an ambitious scientific adventure from Iceland across to Greenland.
The campaign, overseen by the World Meteorological Organization,
is deploying a level of equipment that has never been used in this region before.
The Alliance, a hundred meter long oceanographic boat,
will be home to dozens of scientists from around the world.
It has been kitted out with all the latest oceanographical,
acoustic and meteorological technology.
For this mission, the British Antarctic Society is using Alfa Zulu,
an aircraft specially designed to fly in polar regions.
It can measure wind and turbulence, sea temperature,
particles in the clouds and the sunlight filtering through the atmosphere.
The main goals of the project are to find out more about the climate system
in the Iceland and Greenland seas.
So this is a region where the atmosphere and the ocean are tightly coupled.
They talk to each other.
This is an important component of the main ocean circulation
and key for climate for Northwest Europe.
There's another very important aspect of this,
and that is to help improve the ability to forecast weather.
And, you know, the Arctic is changing drastically.
I mean, incredibly fast, you know.
Yet it's such a data poor region.
So we need data to understand better
how to do things like forecast weather and high latitudes.
And I would contend that not only is it data poor,
but in wintertime it's almost data barren.
Until recently, few scientific programs have been carried out in this part of the world.
The extreme weather here makes observation campaigns particularly challenging,
and it is very difficult to set up instruments for measuring.
Robert Pickhart and Ian Renfrew have had to wait eight months
to begin this project and finally set sail.
Right from the start,
the realities of the Arctic are having a profound effect on everyone.
The long and difficult task of taking measurements and samples begins
as they start trying to uncover the mysteries of the Arctic.
This direction, we're going to get it out, open this flap,
get it out, lock there, and give it to Chris,
and he will let it go.
New tools have been specially developed by the meteorologists.
A prototype submarine drone is being launched for the first time.
It should measure the marine currents it encounters
between the surface of the ocean and a depth of 1,000 meters.
Data like this has never before been captured.
Okay, I'm going to go inside and get on the radio,
and I'll let you know when to re-plug it.
The Arctic currents the drone will measure are key to our terrestrial mechanisms.
They govern the circulation of global ocean currents and thus of our weather.
In Nordic seas, the waters cool under the influence of the atmosphere.
They become denser and descend into abysses and head towards the equator.
As they move south, they mingle with the waters they encounter,
heating up and returning to the surface before heading north again.
This heat redistribution is crucial for the life of our planet.
So this program is looking at one aspect of that big overturning loop.
It's looking at where the water gets dense,
where it sinks, how it sinks, and how it gets into that deep current.
That's what we're looking for,
and that's why we need to be here in winter during the coldest time of the year.
For more than two months,
the Research Vessel Alliance and the Alpha Zulu plane will measure the heat exchanges
that are produced from the depths of the ocean to the highest levels in the atmosphere.
Flight after flight, and from one destination to the next,
the meteorologists hope to find an explanation for the key to changes occurring in the Arctic.
It will take several weeks of very hard work before the Arctic starts to reveal its secrets.
While meteorology researchers take to the field like a new generation of explorers,
weather forecasters are simultaneously developing essential new tools,
without which no predictions could be made.
Since 1975, the European Centre for Medium-Range Weather Forecasts in Reading, UK,
has been revolutionizing the world of weather forecasting, thanks to its supercomputers.
We have about 20 cabinets here, so two times ten cabinets,
and per cabinet we have of the order of 200 compute nodes,
all together roughly 130,000 cores per cluster.
So it's a 260,000 core system that we're running here.
How is it possible to predict weather events in a country or region?
For the meteorologists in Reading,
this involves dividing the globe into segments, like an orange.
They have mapped virtual squares, like the holes in a net.
Initially, these squares were 210 kilometers across,
but a few years later, they were reduced to 125 kilometers, and then 62 kilometers.
The atmosphere was also divided into columns,
and then into horizontal layers, like the floors in a skyscraper.
The more precise this geographical segmentation becomes,
the higher the resolution of the mathematical model,
and the better the predictions for each of these little earthly zones.
So if you translate what we do today, like a nine-kilometer forecast,
we have 137 levels vertically.
Nine kilometers translate to six and a half million grid points,
times 137 levels, times ten prognostic variables.
That translates to something like nine billion updates of variables
every time step we do a calculation.
And with our time step, a 10-day forecast takes about 2,000 time steps,
so you do these updates of nine billion parameters 2,000 times.
It is crazy, but it's the problem we have to solve.
Nowadays, computers are some of the most important tools used by scientists.
The billions of calculations they are able to make
helps better quantify interactions between the Earth, the atmosphere, and the oceans.
But some phenomena are still not very well measured,
which results in lower-quality predictions.
One might think that to do weather forecasting,
you just have to sample the atmosphere,
and the evolution of that atmosphere will be enough.
But the atmosphere interacts with everything, the Earth, the oceans, sea ice,
and even tiny particles of ozone or aerosols in the atmosphere.
And all of that isn't taken into consideration in our models.
So we are driving this concept of coupling,
of interaction between the different components that we call our Earth system.
For example, clouds are one of the phenomena that are so difficult to understand.
They are part of our daily lives,
but are one of the main sources of uncertainty in our forecasting models.
Clouds produce rain.
They give rise to storms.
And they travel great distances.
But the story of these huge fluffy formations,
from their creation to their eventual dissipation, is full of unknowns.
In one of the laboratories of Météo France in Toulouse,
a team is developing a new kind of laser.
These lasers are able to count the number of drops of rain in a cloud,
and detect their volume and concentration.
Cyriel Dangean is particularly interested in very small particles in suspension in the air,
without which there would be no clouds over the Earth, aerosols.
These particles, which range from just a few microns to a few millimeters across,
allow water vapor to transform into droplets.
Aerosols have many sources.
Some are naturally emitted into the atmosphere,
such as desert dust from the great deserts around the world.
Then there are aerosols that come from the oceans,
what we call marine salts.
There are also aerosols that come from plants, which are organic aerosols.
And then there are other aerosols that stem from human activity.
To better understand the role of aerosols,
Cyriel Dangean has had to find a special place
where some very different air masses converge,
bringing a range of different aerosols with them.
She has decided to set up an aerosol trap in the Pyrenees.
At more than 2,800 meters above sea level,
on the summit of the Pic du Midi mountain.
I'm particularly interested in black carbon aerosols
that come from pollution linked to human activity.
Black carbon aerosols are really upsetting our climate.
Been emitted into the atmosphere since the industrial revolution
by factories, airplanes, cars, and wood burned for heating.
And until now, studies of them have been relatively limited.
We are now interested in these particles because we have realized
they may play a critical role in the life cycle of clouds.
If that's the case, then this could be very serious
because black carbon aerosols come from human activity.
And if human emissions increase,
there could be an even greater concentration of black carbon
in the atmosphere which in turn
could change the quantity of clouds in the atmosphere.
But that's not all.
These aerosols also have a secondary effect on the clouds.
They warm the earth by absorbing sunlight.
And this hike in temperature encourages
the evaporation of water droplets in the clouds.
Because of this absorbency,
black carbon is changing the temperature of the atmosphere
and playing a key role in the end of life of clouds
by making them disappear.
The effects of black carbon aerosols around the planet
is still mainly a mystery to us today.
Depending on the season,
clouds are also responsible for catastrophes.
As they develop at the end of the summer,
they often trigger storms.
And these storms, which are often difficult to predict,
sometimes have dramatic consequences.
This bridge in Ville-Gaillac was unable to withstand
the force of the flow.
It has been sliced in two,
and the emergency services have had to evacuate those affected,
like this old lady,
helicoptered in the other side of the river
for treatment to her injured leg.
In the village of Treb,
entire neighborhoods are underwater.
Vehicles have piled up,
roads are impassable,
and houses are destroyed.
We don't know what we're going to do.
It's a disaster.
Why weren't we warned in advance?
Why didn't they tell us to seek shelter?
Two million people around the world now
live in flood zones
and are threatened with the dangers of storms and high water.
Storms have become an essential area of study for meteorologists.
This year in France,
an intrepid team is planning to travel
right to the heart of these inhospitable masses.
Using the meteorological research plane
belonging to Météo France,
the CNRS and the CNES,
a team of around 30 storm hunters,
are planning to study a phenomena
that often occurs with storms,
lightning.
Lightning can happen at speeds
of up to 20,000 kilometers per second.
And measurements have revealed
that the temperature within some of them
can reach up to 30,000 degrees Celsius.
These figures show what is waiting
for Eric Deferre and his team.
We want to understand how storms form
and how lightning forms.
More specifically,
how it is triggered and why it is propagated,
and why it is propagated over such great distances.
Another objective is to look at how data
relating to lightning can be assimilated
into digital weather forecasting models,
and how one day we can use this
to improve our forecasting.
The arrival of storm episodes
over the Mediterranean is the signal
the storm hunters have been waiting for.
Preparations for take-off can begin.
These aren't our normal missions
because usually, as pilots,
we try to avoid storms,
and here they want us to go inside them.
We try to go up as high as possible,
above the anvil of the storm cloud,
and then the roster radar,
which is behind us,
can study the vertical composition of the cloud
and in particular its water,
snow, ice and hail content.
Our objective is to try to get as close as possible
to the storm system,
so we then descend through the cloud
as low as we can,
while ensuring the flight is still safe.
For five hours,
the plane and its team of scientists
pierces the clouds,
aiming for the heart of the storm
and recording the electric fields
that accompany each burst of lightning
with the onboard equipment.
At 8 p.m., they return to the base at Salenzara
for an immediate debriefing.
There are also things we didn't really understand.
An examination of the images
captured by the camera on the plane's nose
has a surprise in store for the scientists.
We're wondering if it wasn't us
that triggered the lightning.
During the flight,
in a fraction of a second,
the plane is hit by lightning.
We can see very well here
that the plane was flying at 8 km of altitude
and the cloud is developing at up to 12.5 to 13 km of altitude,
so we can see that the electric field is increasing
as the plane approaches the center of convection activity
where the lightning occurs.
In the end, for the general public,
it's more about very short-term storm forecasting,
about where a storm is going to be
in 15 or 30 minutes' time
and what shape it will take.
The idea behind the Exedra project
is about providing additional information
to what we already have
to refine the scenario we will be experiencing
in the next 15 to 30 minutes.
In a few years' time,
and thanks to lightning,
meteorologists hope to be able to predict
the location and strength of storms,
not just to the nearest half hour,
but a day before they occur.
Meteorology is a race against the weather,
and with the increase of such devastating phenomena,
it's becoming more essential than ever.
We came running out,
the hurricane was so violent
and it wasn't stopping,
so we went outside.
With every new catastrophe,
the same questions arise.
How can we plan the exact trajectory of hurricanes
and predict how intense they will be?
These phenomena occur almost every year.
Every year since he moved to Martinique,
meteorologist Jean-Noël de Grasse
has seen a dozen tropical cyclones develop
above the Atlantic.
Along with his co-workers,
he establishes the conditions for their formation
off the coast of the African continent.
When the ocean temperature goes above 28 degrees Celsius
and the humidity intensifies,
these turbulent phenomena sometimes turn into cyclones.
On August 30th, 2017,
Jean-Noël watched his 10th tropical cyclone of the year
begin to form.
Irma had all the characteristics of a major hurricane.
The meteorologists were on the highest possible alert.
Hurricane Irma,
rapid intensification,
category two hurricane.
This will be a very dangerous storm
and potentially could affect the Caribbean,
the Lesser Antilles,
Puerto Rico, Dominican Republic,
potentially the East Coast
and the Gulf of Mexico or Florida.
That's why we need to pay attention.
One week before he arrived in the West Indies,
the European Centre for Medium-Range Weather Forecasts
in Reading,
the NOAA in the USA
and Media of France all confirmed that Irma was going to be big.
At that point, the question was
whether it would continue to intensify.
And according to our database, since 1950,
there has never been a category five hurricane
that has hit a small island in the Lesser Antilles.
So the message was,
we don't want to announce a catastrophe,
but this is something quite exceptional.
In the USA,
the National Hurricane Centre
was also monitoring Irma very closely.
Its incredible Hurricane Hunter airplane,
capable of flying into the centre of hurricanes,
took off, heading for Irma.
And right now, I understand,
you're up above this storm, is that right?
Yes, sir.
We're located around the eye of the storm
and what we're doing is we're dropping
what we call dropsondes
and all of that data gets fed into computer models
to help get a better forecast track.
One of the things about this storm
is not only is it extremely intense in the centre,
but it's growing in size.
By early September,
as predicted by the meteorologists,
the diameter of Irma had swelled to an incredible 500 kilometres.
The hurricane was classed as category five,
the highest level,
with winds measuring 300 kilometres per hour.
It was one of the most powerful ever measured above the Atlantic.
Then Irma began to move towards the islands
of Saint Bartholomew and Saint Martin.
The red alert was sounded.
On a personal level,
and I'd even say on a professional one too,
this wasn't something to get excited about.
You know, often meteorologists will say,
wow, that's an amazing hurricane, wow.
That's okay when it's far away,
when it's out at sea.
But this one was a few dozen kilometres from an island,
and we have friends there,
and we've sent people there, willing people,
journalists, people to take measurements.
We were aware of all that,
and so we were wondering if we'd done the right thing,
whether we'd put people in danger.
We were about to witness a humanitarian catastrophe.
That's what we were thinking about at that point.
Despite the accuracy of the weather forecasts given by the models,
the consequences of the passage of Hurricane Irma
were indeed catastrophic.
This hurricane, which also hit the Caribbean and Florida,
resulted in 134 deaths,
the destruction of thousands of buildings,
and cost 65 billion dollars.
Monitoring meteorological phenomena,
understanding where they will strike,
warning populations about the most extreme events.
The responsibility of forecasters are many.
And when the hurricane season ends,
seas that are thousands of kilometres away may also be affected.
When there's a perfect storm between wind, currents, and tides,
the waves striking coastal areas can have devastating consequences
for people, property, and the environment.
They can also result in significant coastal erosion.
Until recently, a handful of satellites
supplied information on oceanic waves using altimeters or radars.
But they didn't allow for a detailed image to be built up.
Nor did they provide precise information
about the relationship between the waves and the currents.
Yes, it's the Talia leaving Brest for a mission on the Iroise Sea.
12 people on board, 3-metre draft, requesting you're OK.
In Brittany, two planes belonging to Météo France and the CNRS,
and one ship from the French Oceanographic Fleet, the Talia,
have been brought in to help better quantify these ocean phenomena.
Fabrice Ardoin is the director of this operation
and works out of the Laboratory for Ocean Physics
and Satellite Remote Sensing in Brest.
His work on developing a way to measure waves on a large scale
has resulted in the newly created SCIM project.
We're in the process of developing a new satellite
that will measure currents and waves.
It uses a Doppler radar, and it's the same principle
as when the police use radars to carry out speed controls.
But instead of measuring the speed of cars,
we're measuring the speed of the surface.
And this visible surface speed provides us with information
about currents and waves.
Using the Doppler effect,
we hope that we'll be able to do it by satellite soon.
Before launching a satellite equipped with this radar,
its effectiveness has to be proven.
The idea is to install it on a plane
and compare the measurements taken from the air
with those taken by the Talia on the sea.
We're leaving in 5 or 10 minutes.
We'll be in the offshore zone for one hour,
and then we'll head for Ouissons to synchronize with the buoys.
They're in position to deploy at midday as planned.
We'll continue with the second one at 20 meters.
Pierre, let me know when you're ready.
I'm ready, Pierre. I'm putting number 10 in the water.
The yellow buoys we just placed in the water
will move up and down with the waves,
and we will measure their movement
the same way as when you move your phone
and it measures the acceleration.
It's the same.
We have a system that measures the acceleration in the buoy,
and using that we can calculate the height,
direction and period of the waves.
Today it's fairly calm, but the swell is quite long,
so we'll be able to check that we can measure
even the smallest amplitude.
We'll check that the quality of the measurements is sufficient,
and that will give us better readings when using the plane.
Having a reliable measurement of ocean currents is key,
and until now we haven't had any satellites
taking such measurements on a global level, like SCIM can.
We've already collected a lot of data
that we've never had access to before.
They are also experts, they have a lot of confidence,
and it's going very well,
so I have a lot of confidence in them as well.
Thanks to SCIM, Fabrice and his team
will be able to measure not just waves,
but also currents that are invisible to the naked eye.
On a planetary scale, this promises to be a colossal task.
Far from Brest, in the Icelandic Sea,
the researchers from the Alliance
are also monitoring ocean currents.
You go to there, and we'll see if we can do a CT there.
OK, so now we are crossing here,
so we can actually eat directly to this one.
You can actually go head to that right there.
Super.
They are trying to understand the weather disturbances
that affect Europe and the USA,
such as cold snaps, summer heat waves and floods.
They've been gathering data in the sea
and in the sky for almost two months now,
and their efforts have been rewarded.
They've just proved the variations in a current
we had no idea existed until a few years ago,
the Northern Icelandic Jet Stream.
So one of the oceanographic goals
is to try and trace the source of that dense water,
whether it's coming from the Iceland Sea,
whether it's coming from the Greenland Sea.
So it's a bit of a puzzle,
and it's kind of trying to trace back
where that water is coming from.
Until now, in terms of the general circulation of the oceans,
scientists had only identified
one single cold current in the Denmark Strait,
the current that runs along the eastern coast of Greenland.
Using temperature, salinity and speed data
gathered by the ship,
they've been able to accurately map
the path of the Northern Icelandic Jet Stream,
a new undersea river.
Today, scientists are putting forward the hypothesis
that the warming of the Arctic
is slowing the flow of this current,
which in turn is altering
the general circulation in the oceans.
These modifications to ocean flows
combined with changing air currents
might help explain the recurrence
of extensive cold periods during wintertime
and heatwaves in the summer.
Travelling to the Arctic, to deserts
or to the middle of the oceans,
crossing through our atmosphere,
gaining a better understanding of clouds,
lightning, waves and currents.
The Earth system is not a simple thing to evaluate.
More than any other science,
meteorology has now become a collaborative affair
across the world
because weather events do not respect borders.
Every two years,
representatives from all 22 EU member states
come together.
Their goal is simple,
to anticipate risk
and ensure the security of populations.
In the scientific world in general,
there is a great deal of cooperation
and in meteorology,
there is perhaps even more than elsewhere.
We collaborate a great deal with our member states,
their meteorological services,
but also their universities and institutions.
We also work with organizations outside Europe
with the USA, with China and with many others.
It's something that really brings together
many of our scientific activities on a global scale.
Major weather research programs
are now decided on a European level.
This was the case this year
with a critically important project called Aeolus,
which will study wind,
a key parameter for weather forecasting
in a way that has never been done before.
In Toulouse,
scientists have designed a satellite
that is capable of measuring wind
in the different layers of the atmosphere
all around the Earth.
Its mission is an audacious one.
Aeolus is going to use
a very high-energy ultraviolet laser
that has never been used before in space.
Agencies like NASA
have already tried to develop similar systems,
without success.
We are going to use the fundamental properties of the laser,
primarily we'll be using the laser's emission frequency.
When the beam strikes particles in the wind,
these particles will send back data that we can use
so we can work out how fast they're traveling,
which will give us a good idea of wind speed.
The only problem is
that when you put an optical instrument on a rocket,
during take-off, the rocket vibrates and shakes.
The impact is like a massive earthquake
and the equipment has to survive that shock.
If it shifts by a single centimeter during take-off,
all our measurements will be out.
And then there's a second challenge.
The sun heats up the sides of the satellite
and there's outer space on the other side,
which is at minus 270 degrees.
So the sun is roasting one side
and it's minus 270 on the other
and the optical instrument needs to be stabilized at 20 degrees
to an accuracy of a few tenths of a degree.
The final problem to resolve is the space radiation
that will attack the equipment's lenses.
The scientists have had to develop special treatments
to protect the laser's lenses against radiation damage.
This project has involved years of testing.
In total, it required 15 years of work
before the satellite was finally ready to launch.
August 22nd, 2018 in Darmstadt, Germany.
This is a key moment.
All space and weather agencies are focused on Kourou,
where the launcher is due to take off,
and on Darmstadt, from where the satellite
will be controlled when it enters orbit.
This is an extremely important moment
because any launch can fail
and that would be 20 years of our lives up in smoke.
Or it could be a great success
and then it's like a baby being born.
It'll be over, we'll have cut the cord
and it'll grow up with us scientists
and that's when we'll see if the mission
will bring everything we hope for.
Top page 0-4 minutes.
Top page 0-1 minutes.
12 of DDO, attention for the final countdown.
10, 9, 8, 7, 6, 5, 4, 3, 2, 1, top.
Ignition, P-80 and take-off, VV-12, Aeolus.
It has to work, it has to work.
Will Aeolus give us a sign of life?
A few minutes after leaving the Earth's surface,
the satellite should now be able to communicate
with the people that launched it.
Yes, it's communicating.
There's a signal and a second one.
It works, it cried.
We can see two peaks
and those are the first cries of our baby.
They look happy, they look very happy.
Soon, observation satellites and ground stations
will be transmitting billions of pieces of data every second.
The future of weather forecasting is being built
in this exponential growth of new equipment,
observations and information.
It is thanks to information technology
that this race is being won.
And yet today, the technical constraints
imposed by the laws of physics
still seem insurmountable to scientists.
Since our strategic goal is to run
five kilometer global ensembles in 2025,
this translates to an inflation
of our computing requirements
by at least a factor of 100.
So our current system is able to run eight petaflops.
This is eight times 10 to the power of 15
floating point operations per second.
This is a very large number.
But if you multiply this by 100,
which would be the requirement
that we would have to fulfill in 2025,
this is basically unaffordable
in terms of computing technology
that you would have to buy
and the electric power to supply it.
We really need to invest
in different types of thinking.
The future of weather forecasting
will involve a technical and scientific evolution,
the effects of which
we are as yet unable to measure fully.
Will forecasters one day manage
to totally model our Earth system
with its winds, clouds, waves and currents?
Will they soon be able to predict the weather
with a resolution of one kilometer
over 200 levels of atmosphere,
10 days in advance as they hope?
Thanks to their research,
the satellites of the future
and all the equipment yet to come,
meteorologists are on the brink
of radically transforming our lives
on a daily basis.
