hello, i'm christian fraser. you're watching the context on bbc news. on ai decoded, we're looking at how artificial intelligence is making weather apps more accurate. experts say it's bringing about a forecasting revolution that could save lives in danger from extreme climate events. if only we could forecast what is happening in the sport. but jane can. hello, jane. maybe not forecast, but i can tell you the latest. international football is back with the nations league under way tonight. scotland are looking for an improved performance in their match against poland at hampden. it's their first competitive game after their early exit from the euros in germany. well, it's not going the way of the home side. they went a goal down in just the eighth minute. scott mctominey then thought he had equalised, but it was ruled out for a handball. however, northern ireland are beating luxembourg 2—0 — paddy mcnair getting the first and dan ballard with the second.

there are nine games going on across three groups tonight. for all the latest, head to the bbc sport website. to cricket and england head coach brendon mccullum has been setting out his expectations as he prepares to take over all of england's cricket teams. he's extended his contract by two years to take control of both the test and white—ball squads. this week, he'll be coaching the test team at the oval, where the series against sri lanka will conclude, and says the timing is now right to take on the two jobs. over the last two years, it would have been nigh on impossible for anyone to do all three formats, and i think hence why, you know, we went down the split coaches route. now, with this, with the schedule easing, i would say slightly easing, but easing enough, i think it gives you the ability to be able to have one person in the role. just another line of cricket to bring you, also involving england, but for their white—ball squad — the captain of which, jos buttler, has been ruled out of the upcoming three—match t20

series against australia, starting next week. he's now also a doubt for the five one—day internationals that follow. buttler, who's 33, hasn't played since the t20 world cup injune because of a calf injury. it's been announced that phil salt will step in to captain england for the first time in the t20s, whilst harry brook is in line to step up, if needed, in the odis. world athletics has paid tribute to the ugandan marathon runner rebecca cheptegei, who's died at the age of 33, saying "she was a talented athlete with lots left to give". cheptegei died in hospital after being brutally attacked by a former boyfriend on sunday. world athletics president sebastian coe said, "our sport has lost a talented athlete", adding...

and the medals continue to stack up for paralympics gb in paris. becky redfern made it third time lucky with gold in the sbi3 100m breaststroke final. redfern, who is visually impaired, had finished second in the event at the past two paralympics, but is the current world champion. the 24—year—old turned well ahead of her rivals and controlled her pace to win in one minute 16.02 seconds. the youngest member of the paralympics gb team, 13—year—old iona winnifrith, has won silver in the pool. the teenager was just a few seconds behind gold medallist mariia pavlova, who finished with a world record time. winnifrith battled hard to take second place in the sb7 100m breaststroke. and on the track britain's sammi kinghorn won herfourth medal of games. she took silver in the t53 400m, less than 2a hours after winning gold in the 100

metres last night. alice tai has picked up her second gold and herfourth medal of the games. she recovered from a bad start in the women's s8 50m freestyle to take the lead in the final ten metres, and then held on to win by half a body length. and that's all the sport for now. —— poland have gone 2—0 up against scotland. it was robert leven off ski with a penalty. you can keep up—to—date with all of that on the sport website. you are watching the context. we are back with our weekly segment, ai decoded. welcome to the programme. we have had a summer break from al decoded. but if, like me, you were on the british beaches, sheltering from the rain, then maybe you were scanning your mobile weather app to see if the sun might ever reappear — which got us thinking, what about al and the weather? how do you predict climate when it's changing so fast? how do you process

the incredible amount of computerised data that is now being generated? well, you model it, and that is where ai is making huge advances. there is a forecasting revolution under way — 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 will hear from the team at oxford university, who are filling in the gaps with al and making it more readily available through cloud computing. or 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 as ever, our regular ai commentator and

and colleague stephanie hare. also here in the studio, the very well known meteorologist florence rabier, director—general of the european centre for medium—range weather forecasts. and joining us on zoom, professor stephen belcher, who's chief of science and technology at the uk met office. welcome to you all. florence, we are going to start you and the earth's digital twin that your team has built in collaboration with the ai industry. let's give the viewers an idea of what it entails, and we can talk off the back.

while. in simple terms, you are stimulating the natural phenomenon —— wow. what, putting it onto a digital twin of the earth?— putting it onto a digital twin of the earth? yes, so a digital oint of of the earth? yes, so a digital point of the — of the earth? yes, so a digital point of the earth _ of the earth? yes, so a digital point of the earth because - of the earth? yes, so a digital| point of the earth because this is a model of everything that the earth is doing that we can

predict through computing occasions. it's a model where all of our knowledge of the physics of the atmosphere in the earth system is encompassed in that model. if the computer programme that we put on a supercomputer and we run it. but all of our knowledge is there about gravity, condensation, storms, etc. if the digital twin of the earth because it's very accurate and it has a very high resolution and its interactive. you can play a bit with it and simulate what if scenarios. i play a bit with it and simulate what if scenarios.— play a bit with it and simulate what if scenarios. i imagine in times gone — what if scenarios. i imagine in times gone by. _ what if scenarios. i imagine in times gone by, you _ what if scenarios. i imagine in times gone by, you did - what if scenarios. i imagine in times gone by, you did that l what if scenarios. i imagine in| times gone by, you did that at a local level, but we're all interconnected, the world is a global environment. our weather systems and activities are all connected. how does this enable you to improve the forecasting that you do? we you to improve the forecasting that you do?—

that you do? we do run these ulobal that you do? we do run these global models, _ that you do? we do run these global models, so _ that you do? we do run these global models, so it's - that you do? we do run these global models, so it's really l global models, so it's really across the whole world. 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 olympic —— atlantic and even the pacific. it's all interconnected, you're right, you have to start global and then refine at the local scale as well. but you really have to know whatever happens around the world at any time in order to go further, and that's what we've been doing for about 50 years in collaborations with our member states. 35 countries in europe supporting this. florence, is this uniquely european, oraround florence, is this uniquely european, or around the world? there _ european, or around the world? there are — european, or around the world? there are several models are the world. because we are european, we are working in collaboration with 35 meteorological services in europe stuff.— meteorological services in europe stuff. are you using historical— europe stuff. are you using historical or _ europe stuff. are you using historical or live _ europe stuff. are you using historical or live data - europe stuff. are you using historical or live data or- historical or live data or

both? _ historical or live data or both? �* ., historical or live data or both? , ., , . both? both, usually the current data for the _ both? both, usually the current data for the weather _ both? both, usually the current data for the weather forecast. | data for the weather forecast. the model has seen the data before the first 12 hours. we go forward and combine again. the latest forecast is the latest data. historically, we've used the data from the last decade and rolling like this. ~ �* ., ., ., this. we've got a real-life example. _ this. we've got a real-life example. i _ this. we've got a real-life example, i think - this. we've got a real-life example, i think the - this. we've got a real-life. example, i think the recent typhoon, typhoon gaemi. one of the other lines that we're looking at?— the other lines that we're looking at? that's typically what we do _ looking at? that's typically what we do is _ looking at? that's typically what we do is a _ looking at? that's typically what we do is a forecast, l looking at? that's typically i what we do is a forecast, this typhoon had dire consequences with 100 people dead and millions affected.

we predict the probabilities for what the weather will do, and all the tactical tracks we think the typhoon will take. which one is the ai model? the re line which one is the ai model? the grey line is _ which one is the ai model? tie: grey line is physics —based which one is the ai model? ti2 grey line is physics —based and the black line is the real observed tract. the blue, what you have is the best estimate that we had before ai. that red one is the _ that we had before ai. that red one is the ai? _ that we had before ai. that red one is the al? the _ that we had before ai. that red one is the al? the ai... - that we had before ai. that red one is the al? the al... that'sl one is the al? the ai. .. that's almost real — one is the al? the ai. .. that's almost real time. _ one is the al? the ai. .. that's almost real time. it _ one is the al? the ai. .. that's almost real time. it is - almost real time. it is incredible, but you can'tjudge everything on one case. it's true that they are about 25% more accurate.

in particular, in terms of the typhoon, 20% worse i know you grew up in tornado valley. that is a made _ grew up in tornado valley. that is a made westford. .. - grew up in tornado valley. that is a made westford... north dakota — is a made westford... north dakota all the way down to texas _ dakota all the way down to texas i_ dakota all the way down to texas. i grew upjust outside chicago, _ texas. i grew upjust outside chicago, and routinely, we practice _ chicago, and routinely, we practice these drills as children. you get a little bit of warning, we're talking seconds, to find the nearest abatement and get underground. presumably, this could tell you what street to go to. hot presumably, this could tell you what street to go to.— what street to go to. not at that sort — what street to go to. not at that sort of _ what street to go to. not at that sort of scale! - what street to go to. not at that sort of scale! i - what street to go to. not at that sort of scale! i have . what street to go to. not at | that sort of scale! i have big expectations, _ that sort of scale! i have big expectations, florence, - that sort of scale! i have big expectations, florence, but| expectations, florence, but it's getting better and better all the time. it it's getting better and better all the time.— let's speak to professor stephen belcher, who's chief of science and technology at the uk met office. welcome to you. so far, we've

talked about weather patterns globally — climate change, mapping, evolution patterns. but how much more precise is it getting day—to—day due to technology? getting day-to-day due to technology?— technology? it's worth remembering - technology? it's worth remembering that - technology? it's worth i remembering that there's technology? it's worth _ remembering that there's always pressure — remembering that there's always pressure to— remembering that there's always pressure to increase _ remembering that there's always pressure to increase the - pressure to increase the accuracy _ pressure to increase the accuracy and _ pressure to increase the accuracy and utility - pressure to increase the accuracy and utility of i pressure to increase the - accuracy and utility of weather forecasts _ accuracy and utility of weather forecasts. today's _ accuracy and utility of weather forecasts. today's great - forecasts. today's great example _ forecasts. today's great example was _ forecasts. today's great example was lots - forecasts. today's great example was lots of- forecasts. today's great| example was lots of rain forecasts. today's great - example was lots of rain here in the — example was lots of rain here in the southwest _ example was lots of rain here in the southwest of— example was lots of rain here in the southwest of england. i in the southwest of england. with— in the southwest of england. with climate _ in the southwest of england. with climate change - in the southwest of england. with climate change making| with climate change making extreme _ with climate change making extreme events _ with climate change making extreme events even - with climate change making extreme events even morel extreme events even more extreme, — extreme events even more extreme, we're _ extreme events even morej extreme, we're demanding extreme events even more - extreme, we're demanding that our forecasts _ extreme, we're demanding that our forecasts get— extreme, we're demanding that our forecasts get better - extreme, we're demanding that our forecasts get better to - our forecasts get better to help — our forecasts get better to help us— our forecasts get better to help us understand - our forecasts get better to help us understand what l our forecasts get better to . help us understand what the impact — help us understand what the impact might— help us understand what the impact might be. _ help us understand what the impact might be. also, - help us understand what the impact might be. also, we l help us understand what the . impact might be. also, we have new applications. _ impact might be. also, we have new applications. just _ impact might be. also, we have new applications. just think - new applications. just think about— new applications. just think about the _ new applications. just think about the roll—out - new applications. just think about the roll—out of- about the roll—out of renewables. - about the roll—out of renewables. this - about the roll—out of. renewables. this means about the roll—out of- renewables. this means the weather— renewables. this means the weather is _ renewables. this means the weather is the _ renewables. this means the weather is the fuel- renewables. this means the weather is the fuel of- renewables. this means the weather is the fuel of the i weather is the fuel of the future _ weather is the fuel of the future. understanding. weather is the fuel of the i future. understanding that weather is the fuel of the - future. understanding that fuel is another— future. understanding that fuel is another application - future. understanding that fuel is another application of - future. understanding that fuel is another application of our. is another application of our weather— is another application of our weather forecasts. - is another application of our weather forecasts. to- is another application of our weather forecasts. to make j is another application of our - weather forecasts. to make them more _ weather forecasts. to make them more accurate, _ weather forecasts. to make them more accurate, we _ weather forecasts. to make them more accurate, we need - more accurate, we need increased _ more accurate, we need increased lead - more accurate, we need increased lead time, - more accurate, we need - increased lead time, warnings further— increased lead time, warnings further ahead _ increased lead time, warnings further ahead of— increased lead time, warnings further ahead of these - increased lead time, warnings| further ahead of these events, but also — further ahead of these events, but also finer—

further ahead of these events, but also finer detail— further ahead of these events, but also finer detail — - further ahead of these events, but also finer detail — justice . but also finer detail — justice france — but also finer detail — justice france describing earlier. i but also finer detail — justice france describing earlier. -- as florence. _ france describing earlier. -- as florence. i'm _ france describing earlier. as florence. i'm going to france describing earlier.= as florence. i'm going to put up as florence. i'm going to put up an image about how accurate this can get. here's an image of london you can all recognise. you can see the dome there. tell us what we're working on the left and the right. fit we're working on the left and the riuht. �* we're working on the left and the right-— the right. at the met office, we complement _ the right. at the met office, we complement what's - the right. at the met office, j we complement what's done the right. at the met office, i we complement what's done in florence's _ we complement what's done in florence's centre _ we complement what's done in florence's centre by— we complement what's done in florence's centre by producing | florence's centre by producing high _ florence's centre by producing high resolution, _ florence's centre by producing | high resolution, physics—based modelling _ high resolution, physics—based modelling of— high resolution, physics—based modelling of the _ high resolution, physics—based modelling of the weather - high resolution, physics—based modelling of the weather of. high resolution, physics—basedl modelling of the weather of the uk. modelling of the weather of the uk the — modelling of the weather of the uk. the left—hand _ modelling of the weather of the uk. the left—hand side - modelling of the weather of the uk. the left—hand side is- uk. the left—hand side is showing _ uk. the left—hand side is showing you _ uk. the left—hand side is showing you the - uk. the left—hand side is showing you the grade i uk. the left—hand side is. showing you the grade that uk. the left—hand side is- showing you the grade that we'd to london— showing you the grade that we'd to london into. _ showing you the grade that we'd to london into. you _ showing you the grade that we'd to london into. you —— - showing you the grade that we'd to london into. you —— divided l to london into. you —— divided london — to london into. you —— divided london. about— to london into. you —— divided london. about one _ to london into. you —— divided london. about one and - to london into. you —— divided london. about one and a - to london into. you —— divided london. about one and a halfl london. about one and a half kilometres, _ london. about one and a half kilometres, we _ london. about one and a half kilometres, we implement i london. about one and a halfl kilometres, we implement the differences— kilometres, we implement the differences in— kilometres, we implement the differences in the _ kilometres, we implement the differences in the rainfall- kilometres, we implement the differences in the rainfall and i differences in the rainfall and the temperatures. _ differences in the rainfall and the temperatures. what - differences in the rainfall and i the temperatures. what we've been _ the temperatures. what we've been able _ the temperatures. what we've been able to— the temperatures. what we've been able to do— the temperatures. what we've been able to do in— the temperatures. what we've been able to do in fact, - the temperatures. what we've been able to do in fact, one i the temperatures. what we've been able to do in fact, one ofj been able to do in fact, one of our rising _

been able to do in fact, one of our rising stars, _ been able to do in fact, one of our rising stars, has— been able to do in fact, one of our rising stars, has devised l our rising stars, has devised machine _ our rising stars, has devised machine learning _ our rising stars, has devisedj machine learning techniques our rising stars, has devised i machine learning techniques to add fine detail— machine learning techniques to add fine detail onto _ machine learning techniques to add fine detail onto those - add fine detail onto those routine _ add fine detail onto those routine forecasts - add fine detail onto those routine forecasts that - add fine detail onto those i routine forecasts that gives add fine detail onto those - routine forecasts that gives us detail— routine forecasts that gives us detail that _ routine forecasts that gives us detail that resolutions - routine forecasts that gives us detail that resolutions of - detail that resolutions of hundreds— detail that resolutions of hundreds of— detail that resolutions of hundreds of metres - detail that resolutions of hundreds of metres in i detail that resolutions of. hundreds of metres in the temperature... _ hundreds of metres in the temperature. . ._ hundreds of metres in the temperature... hundreds of metres in the tem erature. .. ., .., temperature... you can tell the temperature — temperature... you can tell the temperature in _ temperature... you can tell the temperature in the _ temperature... you can tell the temperature in the heat - temperature... you can tell the temperature in the heat in - temperature... you can tell the temperature in the heat in the l temperature in the heat in the rainfall literally over the dome. , ., ., , dome. the temperature at this sta . e dome. the temperature at this stage and _ dome. the temperature at this stage and may _ dome. the temperature at this stage and may be _ dome. the temperature at this stage and may be other- stage and may be other variables— stage and may be other variables in— stage and may be other variables in the - stage and may be other variables in the future, | stage and may be other. variables in the future, but the — variables in the future, but the temperature _ variables in the future, but the temperature is- variables in the future, but the temperature is what i variables in the future, but. the temperature is what lewis and his— the temperature is what lewis and his colleagues _ the temperature is what lewis and his colleagues worked - the temperature is what lewis and his colleagues worked on. and the — and his colleagues worked on. and the reason _ and his colleagues worked on. and the reason this _ and his colleagues worked on. and the reason this is - and the reason this is important _ and the reason this is important is - and the reason this is important is that - and the reason this is. important is that we've and the reason this is - important is that we've known for some _ important is that we've known for some time _ important is that we've known for some time that _ important is that we've known for some time that we - important is that we've known for some time that we have i important is that we've known i for some time that we have heat waves, _ for some time that we have heat waves, they— for some time that we have heat waves, they are _ for some time that we have heat waves, they are more _ for some time that we have heat waves, they are more extreme l for some time that we have heatl waves, they are more extreme in urban— waves, they are more extreme in urban areas _ waves, they are more extreme in urbanareas in— waves, they are more extreme in urban areas. in particular, - urban areas. in particular, those _ urban areas. in particular, those who— urban areas. in particular, those who live _ urban areas. in particular, those who live in - urban areas. in particular, those who live in cities - urban areas. in particular, | those who live in cities will know — those who live in cities will know that _ those who live in cities will know that they _ those who live in cities will know that they don't - those who live in cities will know that they don't cool. those who live in cities will - know that they don't cool down so much — know that they don't cool down so much at _ know that they don't cool down so much at night _ know that they don't cool down so much at night and _ know that they don't cool down so much at night and we - know that they don't cool down so much at night and we call. know that they don't cool down so much at night and we call it| so much at night and we call it the urban— so much at night and we call it the urban heat— so much at night and we call it the urban heat island. - so much at night and we call it the urban heat island. what i the urban heat island. what lewis — the urban heat island. what lewis and _ the urban heat island. what lewis and his _ the urban heat island. what lewis and his team - the urban heat island. what lewis and his team did - the urban heat island. what lewis and his team did wasl lewis and his team did was taking _ lewis and his team did was taking data _ lewis and his team did was taking data from _ lewis and his team did was taking data from crowd - lewis and his team did was i taking data from crowd source data _ taking data from crowd source data in — taking data from crowd source data in back _ taking data from crowd source data in back gardens - taking data from crowd source data in back gardens and - data in back gardens and citizens _ data in back gardens and citizens in— data in back gardens and citizens in london. -

data in back gardens and i citizens in london. variable quality. _ citizens in london. variable quality. plus— citizens in london. variable quality, plus five _ citizens in london. variable . quality, plus five professional whether— quality, plus five professional whether state _ quality, plus five professional whether state —— _ quality, plus five professional whether state —— weather- whether state — — weather stations. _ whether state —— weather stations, they _ whether state —— weather| stations, they augmented whether state —— weather- stations, they augmented the regular— stations, they augmented the regular forecast— stations, they augmented the regular forecast that - stations, they augmented the regular forecast that we - regular forecast that we produce _ regular forecast that we produce here _ regular forecast that we produce here at - regular forecast that we produce here at the - regular forecast that we | produce here at the met regular forecast that we - produce here at the met office and produce _ produce here at the met office and produce these _ produce here at the met office and produce these forecasts . produce here at the met officej and produce these forecasts at and produce these forecasts at a very— and produce these forecasts at a very fine _ and produce these forecasts at a very fine level. _ and produce these forecasts at a very fine level. it's _ and produce these forecasts at a very fine level. it's another. a very fine level. it's another example _ a very fine level. it's another example of— a very fine level. it's another example of how _ a very fine level. it's another example of how ai _ a very fine level. it's another example of how ai can - a very fine level. it's another example of how ai can really change — example of how ai can really change what _ example of how ai can really change what we're _ example of how ai can really change what we're doing - example of how ai can really change what we're doing in. example of how ai can really . change what we're doing in the weather— change what we're doing in the weather world. _ change what we're doing in the weatherworld. [— change what we're doing in the weather world.— weather world. i have a pract on my question _ weather world. i have a pract on my question for _ weather world. i have a pract on my question for you. - on my question for you. sometimes when i'm standing in london, i will consult the met office app and it will say that it's sunny, and i'm being rained on. why is that happening? second, how orwhen willi happening? second, how orwhen will i be able to send data to you saying no, it's raining here? ., , ., ., ., , here? you can send data to us riaht here? you can send data to us right now- _ here? you can send data to us right now. it's _ here? you can send data to us right now. it's called _ here? you can send data to us right now. it's called weatherl right now. it's called weather on the — right now. it's called weather on the web— right now. it's called weather on the web, _ right now. it's called weather on the web, the _ right now. it's called weather on the web, the wow- right now. it's called weather on the web, the wow site,| right now. it's called weather. on the web, the wow site, so please — on the web, the wow site, so please do — on the web, the wow site, so please do that. _ on the web, the wow site, so please do that. that's - on the web, the wow site, so please do that. that's the - on the web, the wow site, so

please do that. that's the data| please do that. that's the data that lewis _ please do that. that's the data that lewis and _ please do that. that's the data that lewis and his _ please do that. that's the data that lewis and his colleagues i that lewis and his colleagues used — that lewis and his colleagues used to— that lewis and his colleagues used to produce _ that lewis and his colleagues used to produce this. - that lewis and his colleagues used to produce this. i- that lewis and his colleagues used to produce this. i thinkl used to produce this. i think in terms _ used to produce this. i think in terms of— used to produce this. i think in terms of weather - used to produce this. i think. in terms of weather forecasts, lets— in terms of weather forecasts, let's not— in terms of weather forecasts, let's not forget _ in terms of weather forecasts, let's not forget over _ in terms of weather forecasts, let's not forget over the - in terms of weather forecasts, let's not forget over the last . let's not forget over the last 50 years. _ let's not forget over the last 50 years, through— let's not forget over the last 50 years, through the - let's not forget over the lastl 50 years, through the advent let's not forget over the last - 50 years, through the advent of satellite — 50 years, through the advent of satellite observations _ 50 years, through the advent of satellite observations and - satellite observations and others _ satellite observations and others around _ satellite observations and others around the - satellite observations and others around the world, i satellite observations and i others around the world, the improvement_ others around the world, the improvement of— others around the world, the improvement of the - others around the world, the improvement of the modelsl others around the world, the - improvement of the models that florence — improvement of the models that florence was _ improvement of the models that florence was describing - improvement of the models that florence was describing in - improvement of the models that florence was describing in the l florence was describing in the increased _ florence was describing in the increased scale _ florence was describing in the increased scale of— increased scale of supercomputers i increased scale of. supercomputers that increased scale of- supercomputers that we've increased scale of— supercomputers that we've got, these — supercomputers that we've got, these physics—based _ supercomputers that we've got, these physics—based models- these physics—based models we've — these physics—based models we've got— these physics—based models we've got improved - these physics—based models- we've got improved tremendously welt _ we've got improved tremendously wetl we — we've got improved tremendously well. we describe _ we've got improved tremendously well. we describe the _ we've got improved tremendously well. we describe the weather- well. we describe the weather forecast — well. we describe the weather forecast improves _ well. we describe the weather forecast improves one - well. we describe the weather forecast improves one day- well. we describe the weather forecast improves one day for| well. we describe the weatherl forecast improves one day for a decade — forecast improves one day for a decade so— forecast improves one day for a decade. so the _ forecast improves one day for a decade. so the four—day- decade. so the four—day forecast _ decade. so the four—day forecast now— decade. so the four—day forecast now is - decade. so the four—day forecast now is as - decade. so the four—day forecast now is as good i decade. so the four—day. forecast now is as good as decade. so the four—day- forecast now is as good as the three-day— forecast now is as good as the three—day forecast _ forecast now is as good as the three—day forecast ten - forecast now is as good as the three—day forecast ten years i three—day forecast ten years ago — three—day forecast ten years ago this _ three—day forecast ten years ago this is _ three—day forecast ten years ago. this is often _ three—day forecast ten years ago. this is often referred . three—day forecast ten yearsj ago. this is often referred to as the — ago. this is often referred to as the quiet— ago. this is often referred to as the quiet revolution. - ago. this is often referred tol as the quiet revolution. what ai as the quiet revolution. what ai is— as the quiet revolution. what al is doing _ as the quiet revolution. what ai is doing is— as the quiet revolution. what ai is doing is really— ai is doing is really exhilarating - ai is doing is really exhilarating that. ai is doing is really- exhilarating that revolution. it's exhilarating that revolution. it's attowed _ exhilarating that revolution. it's allowed revolution. - we're going to see how this can be applied around the world. coming up after the break, we'll bring in the climate

physicist who had just... stay with us.

welcome back. we are warned repeatedly that climate change will affect millions of people worldwide. in fact, it's already affecting lives and livelihoods, and particularly so in some of the poorest regions of the world where they don't have access to real—time forecasting, or the vast computer power needed to produce it. dr shruti nath is a climate scientist at oxford university. she's been working with the un world food programme to develop an ai system that is pulling together all the data, current and historic, and applying that to localised areas. 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 dr shruti nath. thank you very

much for coming on the programme. there were some brilliant physicists like you in the oxford university department. i want to better understand what the ai is doing to speed up the process of filling the gaps.— to speed up the process of filling the gaps. thank you for havin: filling the gaps. thank you for having me- — filling the gaps. thank you for having me. at _ filling the gaps. thank you for having me. at oxford - filling the gaps. thank you for| having me. at oxford physics, we're exploring hybrid modelling approaches, so looking at how ai can best complement our existing weather models. as florence says, these have all the physical knowledge that we've accumulated since newton. we're complement and went with al, particularly at oxford. we're looking at rainfall particularly, and what we're seeing is that when we take the best of the physical weather forecasts, take the best of the physical weatherforecasts, we take the best of the physical weather forecasts, we can really use ai as a data technique to create these

structures that could arrive from incomplete representations to deliver more accurate rainfall forecasts.- to deliver more accurate rainfall forecasts. how do you see ordinary _ rainfall forecasts. how do you see ordinary people _ rainfall forecasts. how do you see ordinary people in - rainfall forecasts. how do you see ordinary people in the - see ordinary people in the regions _ see ordinary people in the regions being able to access this very— regions being able to access this very sophisticated high—power technology that high— power technology that you're — high— power technology that you're working high—power technology that you're working with? high-power technology that you're working with?- high-power technology that you're working with? that's a very good _ you're working with? that's a very good question _ you're working with? that's a very good question and - you're working with? that's a j very good question and that's actually in my opinion one of the real strengths of ai. it's a very low—cost, lightweight model of being able to represent very complex phenomena. so, we were mostly with other local meteorological body —— we work closely, and we developed the model with them. they actually run the ai models in house, and that means they can actually generate weather forecasts on a laptop. mind you, that's a laptop compared to a supercomputer, which is what typically is used to

predict forecasts.- what typically is used to predict forecasts. can you give us a real-time _ predict forecasts. can you give us a real-time example? - predict forecasts. can you give us a real-time example? of. us a real-time example? of course. us a real—time example? of course. we us a real—time example? qt course. we use it in us a real—time example? t>t course. we use it in kenya. they update the forecast every day in kenya and the forecasts are also available on the website. the website name is quac .net, and they're updated every day. it really is a way of giving these people a bit more accessible weather forecasting. more accessible weather forecastina. , , forecasting. presumably, the breakthrough _ forecasting. presumably, the breakthrough of _ forecasting. presumably, the breakthrough of all— forecasting. presumably, the breakthrough of all this - forecasting. presumably, the breakthrough of all this is - breakthrough of all this is that you can, if you know it's coming in the forecast improves, the eight agencies can —— aid agencies can store forward the aid they need that's coming up. so often on our programmes, we say we can't get to these inaccessible areas. but now the stuff will

already be there because we've forecasted what's coming. exactly, so we particularly actually focus on linking research actions, so we work with linking these forecasts to anticipate a reaction. as you said, we can have these long—range forecasts. what florence mentioned about how the weather is chaotic, there's a lot of possibilities that can arise. we have a lot of uncertainty and we need to actually generate a lot of different weather forecasts to explore that. what ai allows you to do this in a very low—cost way. you can generate forecasted that explore the reaction in a... what we haven't talked a lot about is climate change, and of course there are climate deniers out there, florence, that we must acknowledge. i'm going to put a picture on screen. do you remember this?

this was a tornado that was coming up the florida panhandle, and there were questions about whether it might go to alabama and they got a sharpie out and actually drew it on the end, the trump administration. it tells you that it clearly is something that it clearly is something that people try to play with when we talk about climate change and what weather is going to do. you're forecasts are so accurate and presumably with your digital between earth, you can predict how climate will evolve well into the future.— climate will evolve well into the future. , ., , the future. exactly, and we use the future. exactly, and we use the same _ the future. exactly, and we use the same model— the future. exactly, and we use the same model to _ the future. exactly, and we use the same model to do - the future. exactly, and we use the same model to do climate l the same model to do climate models as well. but it's based on the same modelling. also, we can document climate change, and that's what we're doing with the copernicus programme from the eu. going back from 1940 from the eu. going back from 19ll0 and really depicting what the weather and climate have been doing every traveller from

1940 been doing every traveller from 19ll0 until now. —— every hour. we have this picture of the earth and we can document how much the temperature has increased, the frequency of storms, etc. there is already this reality. we have enough information to know what has happened already. with these models, we could do a digital point of the climate as well and go forward in the future. with different scenarios of what will happen in the reduction of greenhouse gases, because it all depends how much we can use the amount of the carbon dioxide in particular. i have a question for all three of us — have a question for all three of us. ~ . . of us. will have a minute let sa ou of us. will have a minute let say you have _ of us. will have a minute let say you have to _ of us. will have a minute let say you have to be - of us. will have a minute let say you have to be quick! is| of us. will have a minute letl say you have to be quick! is it true when _ say you have to be quick! is it true when it _ say you have to be quick! is it true when it comes _ say you have to be quick! is it true when it comes to - say you have to be quick! is it true when it comes to whether ctimate — true when it comes to whether climate change and biodiversity toss. _ climate change and biodiversity toss. we — climate change and biodiversity loss, we have to fix the climate _ loss, we have to fix the climate change and biodiversity probtem~ — climate change and biodiversity problem. do you feel there's a

way for— problem. do you feel there's a way for citizen scientists to .et way for citizen scientists to get back _ 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 — scientists so that you can help us fight these bigger problems? stephen, 30 seconds left. yeah, it's a great _ stephen, 30 seconds left. yeah, it's a great shout _ stephen, 30 seconds left. yeah, it's a great shout and _ stephen, 30 seconds left. yeah, it's a great shout and as - stephen, 30 seconds left. yeah, it's a great shout and as we - it's a great shout and as we talked _ it's a great shout and as we talked about— it's a great shout and as we talked about earlier, - it's a great shout and as we talked about earlier, we - it's a great shout and as we i talked about earlier, we have the weather _ talked about earlier, we have the weather on _ talked about earlier, we have the weather on the _ talked about earlier, we have the weather on the web. - talked about earlier, we have the weather on the web. it'sl the weather on the web. it's another— the weather on the web. it's another great _ the weather on the web. it's another great crowd - the weather on the web. it'sl another great crowd sourcing initiative _ another great crowd sourcing initiative to _ another great crowd sourcing initiative to look _ another great crowd sourcing initiative to look at _ another great crowd sourcing initiative to look at budding i initiative to look at budding and — initiative to look at budding and early— initiative to look at budding and early sighting - initiative to look at budding and early sighting of- initiative to look at buddingl and early sighting of insects around _ and early sighting of insects around the _ and early sighting of insects around the uk, _ and early sighting of insects around the uk, which - and early sighting of insects around the uk, which we've| and early sighting of insects - around the uk, which we've also connected — around the uk, which we've also connected with— around the uk, which we've also connected with climate - around the uk, which we've also connected with climate change l connected with climate change here _ connected with climate change here at — connected with climate change here at the _ connected with climate change here at the met _ connected with climate change here at the met office - connected with climate change here at the met office along. here at the met office along with— here at the met office along with any— here at the met office along with any other— here at the met office along with any other partner- with any other partner organisations. - with any other partner organisations. i- with any other partner organisations. i thinkl with any other partner. organisations. i think it's with any other partner- organisations. i think it's a great — organisations. i think it's a great shout _ organisations. i think it's a great shout for— organisations. i think it's a great shout for citizen - great shout for citizen science, _ great shout for citizen science, this - great shout for citizen science, this one. - great shout for citizen science, this one. i. great shout for citizen science, this one. i could talk claymore. — science, this one. i could talk claymore. as _ science, this one. i could talk claymore. as i _ science, this one. i could talk claymore, as i always - science, this one. i could talk claymore, as i always could. | science, this one. i could talk. claymore, as i always could. we never get to the bottom of everything, but listen, stephen, florence, dr shruti nath, stephanie, thank you. all of our episodes are on youtube. we'll do it again sometime next week. we'll do it again

same time next week. hello there. we've had a real mixture of weather across the uk today — from the dry, sunny weather for west scotland and northern ireland, where there's been barely a cloud in the sky, to the rather different—looking skies across the south of england and wales, with heavy rain moving in here. and, in fact, when the rain arrived across the south of england, it was really intense. had 11l millimetres of rain recorded injust one hour in parts of kent and 12 into parts of hampshire. that's half an inch of rain in the space of an hour. rain that intense, i'm sure, would have led to some localised surface water flooding. and the weather front responsible for that rain is going to stay across those areas all night. the rain, though, is likely to turn at least a little bit lighter for a time. probably quite murky with a lot of low cloud around, some hill fog patches developing here. and across east scotland and eastern areas of england, as the weather turns increasingly humid overnight, there'll be thickening of the low cloud and also a bit of mist, fog and drizzle. so that takes us into friday.

the same weather front is there, same area of low pressure's there. can't move anywhere because of this big area of high pressure blocking its progress, so instead, we'lljust have further prolonged outbreaks of rain across southern areas of england. and although the rain starts off quite light, it'll quite quickly turn heavy with some thunder mixed in. again, there could be some surface water flooding building in. east scotland, eastern areas of england, particularly where you start off with the low cloud, the mist, fog and drizzle, that will eventually clear out of the way for most and it's going to be a warm and humid day. temperatures widely 22—27 degrees celsius, but with further accumulations of rain across the south of england, where we could be looking at some localised flooding issues. now the same low pressure stays with us into the weekend, too, but it is weakening. and so, rather than persistent rain, we're more likely to see showers. those showers heavy and thundery and could pop up just about anywhere across england and wales. scotland and northern ireland, meanwhile, stay dry with some sunny spells. across england and wales. and again, it's the north—west of scotland and northern ireland that will probably stay dry, probably still with

some spells of sunshine coming and going. temperatures still for many of us, reaching the low 20s, which won't feel too bad. we've got another low on the way for next week so it stays unsettled.

hello, i'm christian fraser. you're watching the context on bbc news. hunter biden arrived at court today and everybody expected this to be the jury selection to precede his criminal trial, but quite unexpectedly, his representatives informed the court that he was intending to change his plea. this one that is going to go very long and certainly not before the election, it may even have been next autumn before will be kept to a trial in this case. and, of course, donald trump wins in november, the first thing is going to do is tell the department ofjustice to dismiss the case. joining me tonight are former chair of the democratic committeejess o'connell and leon emirali — pr advisor & former ministerial aide.

first, the latest headlines: