How AI helps predict extreme weather | BBC News
Summary
TLDRIn this episode of AI Decoded, experts discuss the revolutionary impact of AI on weather forecasting. Oxford University's Dr. Shuyi Chen highlights the development of a digital twin of Earth, a supercomputer model that simulates climate change scenarios. The European Center's Florence Rabier and the UK Met Office's Professor Steven Belcher emphasize AI's role in enhancing accuracy, particularly in predicting extreme events. They also touch on citizen science initiatives, encouraging public participation in data collection to combat climate change and improve forecasting.
Takeaways
- 🌐 AI is revolutionizing weather forecasting by processing vast amounts of data and creating highly detailed models of the Earth's climate system.
- 🔍 Oxford University is developing a 'digital twin of the Earth' to simulate natural phenomena and human activities, aiding in the prediction of extreme weather events.
- 🌡️ AI models are improving the accuracy of weather predictions, particularly in tracking tropical cyclones, with an increase of about 25% in accuracy.
- 🌍 The European Center for Medium-Range Weather Forecasts (ECMWF) collaborates with 35 countries to run global models that predict weather patterns worldwide.
- 💻 AI technology allows for the creation of high-resolution, physics-based models that can predict weather down to specific urban areas, such as differentiating temperatures across London.
- 🌪️ AI is particularly useful in predicting severe weather events, which are becoming more frequent and intense due to climate change.
- 🌤️ The UK Met Office is using AI to enhance weather forecasts with fine details, such as temperature variations within urban areas, by incorporating machine learning techniques.
- 🌿 Citizen scientists can contribute to climate research by submitting local weather data, which can be used to improve models and predictions.
- 🌎 AI is being used to not only predict immediate weather but also to model long-term climate change scenarios, helping to plan for the future.
- 🛠️ Despite advancements, AI models are not perfect and have areas for improvement, such as predicting the intensity of typhoons, which is currently less accurate.
Q & A
What is the significance of AI in weather forecasting?
-AI is revolutionizing weather forecasting by processing vast amounts of data and making predictions more accurate. It enables the creation of digital twins, which are highly detailed models of the Earth that can simulate different climate scenarios, helping scientists predict the evolution of climate change.
How does the digital twin of the Earth aid in climate prediction?
-The digital twin of the Earth is a super model that encompasses all knowledge of the physics of the atmosphere and the Earth system. It is run on supercomputers and allows for the simulation of natural phenomena and human activities, providing highly detailed, interactive data that supports decision-making around extreme weather events.
What role does cloud computing play in making AI-generated weather forecasts more accessible?
-Cloud computing allows for the storage and sharing of AI-generated weather forecasts, making them readily available to governments and aid agencies. This technology enables regions with limited resources to access sophisticated weather forecasting tools, which can be run on a laptop rather than requiring a supercomputer.
How has the collaboration between Oxford University and the UN World Food Program improved weather prediction in the Horn of Africa?
-Oxford University has developed an AI system in collaboration with the UN World Food Program that pulls together current and historic data to provide localized weather forecasts. These forecasts are accessible on a website and are generated using AI models that can run on local equipment, enhancing the ability of governments and aid agencies to prepare for climate disasters.
What is the impact of AI on the accuracy of predicting tropical cyclones?
-AI models have shown to be approximately 25% more accurate in predicting the track of tropical cyclones, such as typhoons and hurricanes. However, they are still being improved upon, as they are about 20% less accurate in predicting the intensity of these storms.
How does the UK Met Office use AI to enhance weather forecasting?
-The UK Met Office uses AI to add fine detail onto routine forecasts, providing high-resolution data on temperature and rainfall. This is achieved by combining crowdsourced data with machine learning tools, which allows for the production of very detailed forecasts that can inform anticipatory actions, especially in urban areas where heat islands can exacerbate weather conditions.
What is the 'weather on the web' initiative mentioned in the script?
-The 'weather on the web' initiative is a platform where citizens can submit their local weather observations, such as temperature and rainfall, to contribute to weather forecasting. This crowdsourced data is then used by organizations like the UK Met Office to improve the accuracy and detail of their forecasts.
How has the accuracy of weather forecasts improved over the past decades?
-Over the last 50 years, the accuracy of weather forecasts has improved tremendously due to advancements in satellite observations, global models, and increased computational power of supercomputers. The improvement is often measured by the increase in the forecast range, with the 4-day forecast now as accurate as the 3-day forecast was a decade ago.
What is the potential of citizen scientists in contributing to weather and climate studies?
-Citizen scientists play a crucial role by contributing local data through initiatives like 'weather on the web' and other crowdsourcing platforms. This data helps scientists to better understand and predict weather patterns and climate change, ultimately aiding in the fight against these global challenges.
How does the EU's Copernicus program document and predict climate change?
-The Copernicus program uses historical weather and climate data to document changes in temperature, storm frequency, and other climate indicators. It also uses these models to predict future climate scenarios based on different scenarios of greenhouse gas emissions, particularly carbon dioxide.
Outlines
🌐 Revolutionizing Weather Forecasting with AI
The segment begins with a discussion on the integration of AI in weather forecasting, emphasizing the need for accurate climate predictions amidst rapid environmental changes. AI's role in processing vast amounts of data and modeling the Earth's climate is highlighted, with references to Oxford University's work on AI and cloud computing to make forecasting more accessible. The European Center for Medium-Range Weather Forecasts and the UK Met Office's use of AI to create a 'digital twin' of Earth for detailed climate simulations is explored, showcasing how these technologies can predict climate change and assist in disaster preparedness.
🌍 Global Collaboration in Weather Modeling
This paragraph delves into the global nature of weather forecasting, noting the collaboration between European and American models. It discusses the use of both historical and real-time data in forecasting, with a case study on Typhoon G. The segment highlights the improved accuracy of AI in tracking tropical cyclones by about 25%, while also acknowledging its current shortcomings in predicting storm intensity. The discussion also touches on the personal experiences of a meteorologist growing up in Tornado Alley, emphasizing the potential of AI to enhance warning systems.
🏙️ High-Resolution Weather Modeling and Citizen Science
The focus shifts to high-resolution weather modeling, particularly in urban areas like London. The Met Office's use of AI to enhance local weather forecasts with data from citizen scientists is discussed. The segment explains how machine learning can improve temperature forecasts, especially in urban heat islands. It also addresses the challenge of matching预报 accuracy with localized weather phenomena and how citizen science can contribute to more precise weather predictions.
🌡️ AI in Climate Research and Localized Forecasting
The conversation turns to how AI complements traditional weather models, especially in predicting rainfall—a critical aspect of localized weather. Dr. Nath from Oxford University discusses the development of AI systems that can run on basic hardware, making sophisticated weather forecasting accessible to regions with limited resources. The segment also covers the application of these AI models in Kenya and Ethiopia, providing real-time examples of how they improve weather forecasting and aid in disaster preparedness.
🌌 Documenting Climate Change and Engaging Citizen Scientists
The final paragraph summarizes the program's discussion on climate change documentation and the role of citizen scientists. It covers the EU's Copernicus program's efforts to track climate changes since 1940 and the potential of AI to predict future scenarios based on greenhouse gas emissions. The segment concludes with a call to action for citizen scientists to contribute data, which can help combat climate change and biodiversity loss, highlighting initiatives like the Met Office's 'weather on the web' and other crowdsourcing projects.
Mindmap
Keywords
💡AI
💡Digital Twin
💡Climate Change
💡Weather Forecasting
💡Supercomputers
💡Cloud Computing
💡Machine Learning
💡Citizen Science
💡Extreme Weather Events
💡Renewables
Highlights
AI is revolutionizing weather forecasting by processing vast amounts of data and making predictions more accessible to governments worldwide.
Oxford University's team is using AI to fill gaps in weather data and making it more accessible through cloud computing.
The EU's Destination Earth initiative is creating a digital twin of the planet to predict natural phenomena and climate change evolution.
AI combined with climate science and supercomputers is used to create highly detailed, interactive data for decision-making during extreme weather events.
ECMWF runs global models to predict weather interconnectedness, which is crucial for accurate medium-range forecasts.
AI models have shown to be approximately 25% more accurate in predicting the tracks of tropical cyclones like typhoons and hurricanes.
Despite advancements, AI models still have room for improvement, particularly in predicting the intensity of weather phenomena.
Citizens can contribute to weather forecasting by submitting local data through initiatives like 'weather on the web'.
The UK Met Office is using AI to increase the precision of weather forecasts down to the level of specific landmarks.
AI allows for the generation of multiple weather forecasts to explore uncertainty, aiding in anticipatory actions for climate disasters.
Oxford University is pioneering an AI system that can run on a laptop, making sophisticated weather forecasting accessible to developing regions.
The digital twin of the Earth not only predicts climate change but also documents historical weather and climate data.
Climate change documentation is crucial for understanding historical weather patterns and informing future predictions.
Citizen scientists can play a role in climate change studies by contributing data through crowdsourcing initiatives.
The accuracy of weather forecasts has improved significantly over the past decades, with AI accelerating this progress.
The program concludes with a call for citizen science involvement in climate and biodiversity studies.
Transcripts
we are back with our weekly segment AI
[Music]
decoded welcome to the program we have
had a summer break from AI decoded but
if like me you were on the British
beaches Sheltering from the rain then
maybe you were scaning your mobile
weather app to see if the sun might ever
reappear which got us thinking what
about Ai and the weather how do you
predict climate when it is changing so
fast how do you process that incredible
amount of computerized data that is now
being generated well you model it and
that is where AI is making huge advances
there is a forecasting Revolution
underway so accurate says the guardian
and now in much more accessible format
that very soon governments around the
world will be able to save lives and
protect livelihoods before extreme
events even occur we'll hear from the
team at Oxford University who are
filling in the gaps with AI and making
it more readily available through Cloud
computer shoting or how about this from
the EU destination Earth a digital copy
of our planet on which scientists are
running complex simulations to predict
natural phenomena AI combined with
climate science powered by
supercomputers a digital twin if you
will that will help scientists predict
the evolution of climate change with me
has ever our regular commentator and
colleague uh Stephanie har is here also
in the studio the very well-known
meteorologist Florence rabier Dr rabier
director general of the European Center
for medium-range weather forecast and
joining us also on Zoom Professor Steven
Belcher who is chief of Science and
Technology at the UK Met Office welcome
to you all um fla uh we're going to
start with you um and the Earth's
digital twin that you and your team have
built in collaboration with the the AI
industry so let's get a view for the
viewers let's let's just show the
viewers what it entails and we'll talk
off the
back to create a better future we must
push the boundaries of
today simulations of our Earth system
known as digital twins will help us
understand predict and plan for a
rapidly changing
world this twin will offer highly
detailed Interactive
data that can support decision making
around extreme weather
[Music]
events this twin will show us possible
Futures it simulates different climate
change scenarios over many decades
helping us to be ready for whatever
happens wow so in simple terms you are
stimulating the natural phenomena and
the human activities on Earth putting it
all together through this supercomputer
and what putting it
onto a digital twin of the earth yes so
digital twin of the Earth because it's
this super model of everything that the
Earth is doing that we can predict
through Computing equations so it's a
model where all our knowledge of the
physics of the atmosphere and the Earth
system is encompassed in that model so
it's a computer program that we put on a
supercomputer and we run it but all our
knowledge of the physics accumulated
since Newton and all is there about
gravity condensation you know storms Etc
and it's a digital twin of the Earth
because it's very accurate and it has a
very high resolution and also it's
interactive you can play a bit with it
and simulate what if scenarios so I
imagine that in Times Gone by you would
do that at a very local level but of
course we're all interconnected the
world is is is a global environment our
our weather systems uh and activities
are all connected so so how does this
enable you to improve the forecasting
that you do yeah so at ecmwf what we do
we run these Global model so it's really
across the whole world because if you
want to know the weather in Europe now
you have to know what happened in the US
a few days ago and in the Atlantic and
even in the Pacific if you want to
predict the weather 7 Days beforehand so
it's all interconnected you're right you
have to start with a global scale and
then you can refine at the local scale
as well but you really have to know what
whatever happens on the world at any
point in in time in order to go further
in time in your predictions and that's
what we've been doing for about 50 years
in collaborations with our member states
35 countries in Europe supporting this
work so Florence is this a uniquely
European initiative or do you work with
other partners around the world so there
are several Global models in the world
and our because we are European we are
working in collaboration with 35
meteorological services in broadly
Europe broad Europe but Amer model Etc
ah okay but are you using historical
data or like live data or both probably
both I mean for the weather forecast
usually we use the current data so the
data we've seen in the last 12 hours but
the model had seen the data beforehand
and it's a sort of continuous process we
combine physics and data we go forward
combine again so the latest forecast
uses the latest data but historically
we've used the data from the last
decades and rolling like this I tell you
what we we've got a a real life example
uh that mapped the recent typhoon
typhoon G I think it was called uh in
July this year so so you see all the
lines around the two main lines so
there's the red line and the black line
which we'll talk about in a second what
are the other lines that we're looking
at so that's typically what we do as a
forecast so we we predict the weather
but in particular we concentrate on
severe events like that because this
typhoon G me was really uh had dire
circum you know consequences with 100
people dead and millions affected so
what we produce every day we produce not
just one forecast but we produce
actually several forecast together to to
depict the whole probabilities of what
the weather will do this way we don't
just simulate the track of the typhoon
but all the possible tracks that we
think the typhoon will take in the next
few days these are the gray lines so
which is the AI model
so the gray Line's our physics based
model and the black line is the real
observed track of typhoon and in the
blue what you have is our best estimate
that we had before AI of where the
typhoon would go and would hit China
he's telling me that red one is the AI
that's almost tracking exactly what
happened in real time in that case it is
and that's incredible it is incredible
but you can't judge everything on one
case of course we accumulate all the
these cases and we do statistics but it
is true that the AI models are in
general about 25% more accurate in
predicting the track of tropical
Cyclones typhoon and hurricanes which is
huge but they are not doing everything
right either so in particular in terms
of the intensity of the typhoon they are
actually about 20% worse so it's not all
perfect you've got a real interest in
this because I know that you grew up in
tornado Valley in America right so
Tornado Alley is the midwestern part of
the United States that runs from North
Dakota all the way down to Texas and
then probably for 500 to a, miles on
either side so I grew up just outside
Chicago and routinely we practice these
drills as children you get a little bit
of warning we're talking seconds and you
have to find the nearest basement and
get underground because a storm will
come through that can destroy an entire
town in seconds presum this could tell
you which street to go to well we're not
at that sort of scale especially as we
work I have big
expectations but but it's get but it's
getting better and better all the time
it is getting better all let's bring in
Professor Steven Belcher who is the
chief of Science and Technology at the
UK Met Office um welcome to you um so
far we've talked about global weather
patterns uh Stephen climate Trends
mapping evolution of weather patterns
but how much more precise is weather
prediction getting dayto day because of
this AI
technology well it's worth remember
remembering that there's always pressure
to increase the accuracy and the utility
of weather forecasts uh today's great
example we've had certainly lots of rain
here down in the southwest of England so
with climate change making extreme
events even more extreme we're demanding
that our forecasts get better to uh help
us understand what the impacts of those
might be also we've got new applications
of weather forecasts and just think
about the roll out Renewables this is
meaning that weather is now the fuel of
the future so understanding that fuel is
another application of of our weather
forecasts and so to make them more
accurate we need increased lead time so
we need warnings further ahead of when
we're getting these extreme events but
we'd also like finer detail just as
Florance was describing earlier I was
making quite big demands of Florance but
I'm going to put up an image here about
just how accurate this can get so here's
a here's an image of London that we'll
all recognize um uh you can see the
Millennium D there the bend in the River
East London tell us what we're looking
at here on the left and on the
right right so at the Met Office we
complement What's Done in Florence's
organization the European Center by
producing highresolution physic Bas
modeling of the weather over the UK so
the leftand side is showing you the grid
that we divide London into in order to
provide that weather so what what we're
seeing now is the weather on at that
kind of resolution about 1 and 1 half
kilom we we kind of increment the the
differences in the rainfall and the
temperatures what we've been able to do
in fact one of our Rising Stars Lewis
blun here at the Met Office working with
students at the University of reading
and also at the Bureau of meteorology in
Australia um has devised a machine
learning techniques to add fine detail
onto those routine for CS that gives us
detail resolutions of hundreds of meters
in the temperature so you can tell the
temperature and the Heat and the
rainfall literally over the Millennium
Dome uh the temperature at this stage
and maybe other variables in the future
but but the the temperature is what
Lewis and his colleagues uh worked on
and the reason this is important is that
we we've known for some time that uh
when we have heat waves um those heat
waves are more extreme in urban areas so
in particular those who live in cities
will have noticed that the temperatures
don't cool down so much at night and we
call that the urban heat island so what
Lewis and his team did was to take data
actually from
crowdsourced data in back Gardens and uh
citizens in London their data so of
variable quality frankly plus five
professional weather stations they Mash
that together together with machine
learning tools and augment the regular
forecast that we produce here at the Met
Office and can then produce these
temperature forecasts at these very fine
levels so it's another example of how AI
can really change what we're doing in
the in the weather World Professor beler
I've got a question for you exactly
about this um taking data from your back
Garden sometimes when I'm standing in
London I will consult the Met Office app
regularly in fact and it will say that
it's sunny and I'm being rained on why
is that happening and second how or when
will I be able to send data to you
saying no no here in hatney it's
raining so you can you can send data to
us right now so it's called weather on
the web the wow site so please do that
you can Lodge your longitude and
latitude and send in your data to us and
as I say that's the data that Lewis and
his colleagues used to produce this um I
think
in terms of weather forecast let's not
forget that over the last 50 years
through the Advent of satellite
observations and other observations
right around the world the Improvement
of those global models that Florence was
describing earlier and and the increased
scale of supercomputers that we've got
these physics-based models we've got the
weather forecast has increas improved
tremendously well and and one of the
statistics we describe is that the
weather forecast incre improved by one
day per decade so the 4-day forecast now
is as good as the three-day forecast was
10 years ago so this is often referred
to as the quiet revolution in weather
forecasting what AI is doing is really
accelerating that Revolution so it's a
loud Revolution we're going to continue
the conversation we have to get a short
break but we're going to see how this
can be applied around the world um
coming up after the break we'll bring in
the climate physicist Dr shuy Nath she
is part of the physics team at Oxford
where they just pioneered a new approach
to predicting extreme wither stay with
us welcome back we are warned repeatedly
that climate change will affect millions
of people worldwide in fact is already
affecting lives and livelihoods and
particularly so in some of the poorest
regions of the world where they don't
have access to this realtime forecasting
or the vast computer power needed to
produce it well Dr shuy na is a climate
scientist at Oxford University she's
been working with the UN world food
program to develop an AI system that is
pulling together all this data current
and historic and applying that to
localized area that information can now
be condensed and shared through cloud
computing to help the governments and
Aid agencies better prepare for climate
disasters let's bring in our guest then
Dr shudy D It's good to talk to you
thank you very much for coming on the
program
um there were some brilliant physicists
like you in the in the Oxford University
Department I I want to better understand
that what the AI is doing to speed up
the process and fill in the
gaps yeah so thank you for having me at
oxid physics we're exploring hybrid
modeling approaches so we're looking at
how AI can best complement our existing
physical weather models so as Florence
says these models have all the physical
knowledge that we've accumulated since
Newton and we're complimenting it with
AI particularly at Oxford we're looking
at rainfall since this is a high impact
very localized feature of the weather
and what we're seeing is that when we
take the best of the physical weather
forecasts um we can really use AI as a
data driven technique to correct the
structural errors that exist in these
physical forecasts um that could arise
from incomplete representation of
atmospheric processes to better deliver
actual accurate rainfall forecasts um
within the region that we work in so
that's the greater Horn of
Africa and um Dr n how do you see
Ordinary People in the regions where
you're working being able to access this
very sophisticated and high powerered
technology that you're working
with so that's a very good question and
that's actually in my opinion one of the
real strengths of AI it's a very lowcost
lightweight model of being able to
represent very complex phenomena so we
work closely with all the local
mological bodies and we work with
developing the model with them so they
actually run the AI models um inhouse
and that means that they can actually
generate weather forecasts on a laptop
mind you that's a laptop as compared to
a supercomputer which is what typically
is used to generate weather forecast can
you give us an example of where you've
used that a real a real time example of
course yeah so we use it in Kenya so
they update the forecasts every day um
in Kenya and Ethiopia and the forecasts
are also available on a website so um
the website name is sean. pac. net and
they're updated every day um from the
in-house forecast generated on you know
their uh equipment so it really is a way
of giving these people a bit more access
ible weather forecasting but presumably
that I mean the Breakthrough of all this
is that you can if you know what's
coming and and the long range
forecasting improves you can the aid
agencies can can put can sort of store
forward the aid that they're going to
need for what's coming at them so you
know so often on our programs we're
we're sort of saying well we can't get
to these inaccessible areas but now the
stuff will already be there because
we've already forecasted what's coming
exactly so we particularly actually
focus on linking research to actions so
we work with linking these forecasts to
anticipatory action and as you said we
can have these long range forecasts and
also chipping in on what Florence
mentioned about how you know weather is
quite chaotic so there's lots of
possibilities that can arise from a
given starting point so we have a lot of
uncertainty and you need to actually
generate a lot of different weather
forecasts to explore that AI allows you
to do that in a very lowcost way so you
can generate forecasts that explore the
uncertainty space in a very lowcost
manner so that you can actually properly
inform anticipatory action in these
areas we've talked a lot about weather
what we've not actually talked a lot
about is is climate change and and of
course there are climate deniers out
there Florence uh that we must uh
acknowledge I'm got to put a picture on
screen do you ever remember this uh this
was a a tornado that was coming at the
Florida Panhandle um and also was some
questions about whether it might go to
Alabama and and they got a Sharpie out
and they actually drew it on the end at
the Trump Administration which uh which
tells you that that uh you know we it
clearly is something that that people
try to play with when when we talk about
climate the way climate's changing what
weather is going to do but your your for
are so accurate Florence and presumably
with your digital twin earth You can
predict how climate is going to evolve
well into the future well exactly and we
use the same weather model to to do
climate models as well so they are just
a bit more complicated but it it's based
on the same sort of of modeling but also
we can document climate change and
that's what we're doing with the
Copernicus program from the EU going
back since you know from 1940 and really
depicting what the weather and climate
have been doing every hour from 1940 to
now we have this picture of the Earth
and we can then document how much the
temperature have increased how much the
frequency of storms have increased Etc
so it's predicting it but there is
already this reality we not we have
enough information to know what has had
happened already and then with these
models then we can do a digital twin of
of the climate as well and go forward in
the future with different scenarios of
course of what will happen in the
reduction of greenhouse gases because of
course it all depends how much we can
reduce the amount of carbon dioxide in
particular that we put in the
atmosphere and I've got a question for
all three of our distinguished scientist
only got a minute left so you're going
to have to make it quick go we keep
hearing that we're running out of data
and that this is a big problem for AI
but I wonder if that's actually true
particularly when it comes to weather
climate change and biodiversity loss we
have to fix obviously the climate change
and biodiversity problem do you feel
that there's a way for citizen
scientists to get back into action and
be submitting data to all of you
scientists so that you can help us fight
these bigger problems St Stephen pick
that up because we've just about 30
seconds left yeah it's a great shout and
as we talked about earlier we've got the
weather on the web there's another great
crowdsourcing initiative to look at um
in and early sighting of insects around
the UK which we've also connected with
climate change um here at the Met Office
along with many other partner
organizations so I think it's a great
shout for citizen science this one I
could talk plenty more as I always could
on this program every week we we never
get to the bottom of everything but
listen uh Florence Stephen Dr Nath
Stephanie thank you all very much uh for
your time really fascinating discussion
just a reminder we are putting all these
half hour programs on the BBC's Aid
decoded YouTube
site so you can find all our pass
programs there we'll do it again same
time next week thanks for watching
Ver Más Videos Relacionados
49. Weather Forecasts | The Economics of Everyday Things
What is climate change? - The Climate Question, BBC World Service
PH weather forecasting system, gagamitan na ng artificial intelligence – DOST
Everest Weather - Data is in the Clouds | National Geographic
How AI Can Help and Hurt the Environment | WSJ Tech News Briefing
Previsão do Tempo – Ciências – 8º ano – Ensino Fundamental
5.0 / 5 (0 votes)