UK Weather Risk, Power-System Stress and the Forecast Gap | Rob Hutchinson, Meteomatics

So yeah, quick introduction first, Rob Hutchinson, Meteormatics, for those that don't know we're a weather data and related analytics shop if you like, so yeah I was wondering about the best place to start for this presentation I thought well actually 30 kilometers above our head right now something unprecedented is happening basically a reversal in the stratospheric polar vortex and it's the earliest on record and it's the earliest on record that something like this has happened and this could have some pretty interesting consequences for UK winter and there's a word I'm going to be using a few times during this presentation that's unprecedented it keeps cropping up time and time again in a weather context but then also in related energy markets so I'm going to be taking a look at some of these common themes that we're seeing so reflecting back on summer 2026 unprecedented although we didn't break the all-time UK record 40.3 degrees in a couple of years back what we did see was just sustained heat and repeated heat waves throughout the summer as a whole this put summer 2026 at the 16.5 degrees CET which is a measure of the average temperature across the UK breaking it by 0.4 degrees is a wide margin in climatological terms it's real wide margin but you see it's been broken repeatedly in the last 20 years or so so this is a real trend climate change unfolding before our eyes and this summer and this summer like I say we didn't break the all -time record we came close a couple of times but 54 individual weather stations broke their all-time records so again these records are really tumbling with increasing frequency and it's beginning to have you know it's beginning to put the UK energy system under stress I would say so summer 2026 was it really extreme or is it a new normal I'd say it was probably somewhere in between those two Europe as a whole is warming on average warmer than any other continent and this is because partly where we are quite high latitudes changes in circulation patterns and so on interestingly you see the big cold blob in the North Atlantic this is something we've got to keep an eye on because this could be the start of the Gulf Stream beginning to break down which again is a complicating factor but in summer at least it's one of the reasons it kind of helped to promote a circulation pattern that basically yeah basically it was high pressure across a lot of Europe and interestingly this summer was it was really well signposted all of the teleconnections all of the models that we look at in the longer range they were all pointing towards this summer being being notable not always the case and we'll come back to that in the context of this coming winter um and yeah I think this is a telling stat here I mean you can look back and do these climate attribution studies the Met Office do these things and the stat there is that a summer akin to 2026 is roughly now a one in nine kind of return period event compared with one in 1,000 years pre-warming so that's two orders of magnitude more likely and then I guess symptomatic of the the stress here right I mean electricity margin notices issued by NISO if you could think of those as a symptom of system stress these things you only used to really happen in winter but all of a sudden in this summer we've seen a series of them um five June to September largely related to heat plus some other factors there's usually a kind of combination of different factors here that are at play um a further one in October not really related to cold and then what I didn't add here was that there was one yesterday as well um this was more related to cold dunkel flouter type conditions that have already been mentioned so I think that's that's a symptom right that climate change global warming combined with the you know the the increasing what you know renewable energy sources on the grid these things are coming together and there's more and more kind of system stress that we're beginning to see system stress kind of symptoms um the electricity margin notices but then also volatility in pricing these these kind of things that go hand in hand with scarcity events so I guess in terms of a case study of what one of these system stress events look like and um the important the increasing importance of kind of high performance weather data um is the 9th of October um now an electricity margin notice wasn't actually issued in this event but NISO did have to call on France for emergency assistance and yeah it was a pretty interesting event I mean it was very localized and it was um yeah quite a small scale weather phenomena um I'll go on to it in the next slide but you see this kind of series of events here um basically um about 700 megawatts um all of a sudden came off of two offshore wind farms off the Suffolk coast and you see there that the um the PNs in the light green line um compared to what was metered in the blue line um basically this led to NISO kind of scrambling for capacity and then ultimately having to call on on the interconnector capacity what caused this from a weather perspective um so basically there was this kind of quite a localized squall that passed over these two wind farms roughly at the same time a temporary spike in wind speeds but then there was a lull after the event passed and this just wasn't captured by the type of um global weather models in the middle panel there um and this illustrates why you need a higher performing um weather model to be able to capture such localized small scale weather phenomena that if they if they can be quite consequential for energy forecasting if they tend to pass over clusters of capacity for example so this just really illustrates the increasing importance of high quality um weather forecasts um shameless plug there's the meteomatics forecast on the left ecmwf in the middle panel and then what does this look like from a wind power production forecasting perspective um well there's no such thing as a perfect weather forecast but you've got meteomatics forecast in the turquoise colored line compared to um actual production in the shaded area and what was notified to NISO in the the light dashed line there so it's just a nice case study kind of illustrating where high performance weather data is increasingly required and by high performance I mean um high resolution rapid refresh it's just better able to capture these type of localized phenomena that can be quite consequential so what about the forthcoming winter well it's a complicated picture and I'm going to use the word unprecedented again um basically we've got this um yeah um super el nino that's unfolding at the moment um way over in the tropical pacific and it looks like it's probably going to be the strongest on record there's a little bit of debate about that at the moment but it's going to be a strong one regardless um and there's a little bit of a common misconception for Europe and the UK that el nino automatically equals a stormy mild winter for the UK um and the news story there on the on the right hand side I think that was the UK government got a little bit excited about how wet and stormy and severe weather and so on and warning us to stock up for example um but it's a misconception um it's basically like digging into some of the stats here and we've got the el nino winters on the top line there and other winters on the bottom line and what I attempted to do here is to um yeah analyze or synthesize a kind of stress factor for UK winters over the last um 60 years or so so stress being dunkel flouter essentially so cold still conditions really there's no real strong correlation between el nino and um system stress if you like or kind of stressful weather conditions there's maybe a marginal tilt there but only really marginal and then you get some of those winters that were el nino winters like 2009 2010 um this is a satellite image from that winter um December when the UK was completely blanked in snow so yeah in my opinion anyway this is a bit of a misconception and it's way more complicated than that so yeah I guess in summary El Nino might tilt the balance slightly towards mild conditions but it certainly doesn't rule out cold um or sustained cold and I've attempted to do a similar thing here but with dunkel flouter episodes so we've had the first of these over the last couple of days and it's illustrated why they are increasingly stressful for the UK energy system um NISO issued the first EMN of the winter yesterday and again there's no real correlation between strong El Ninos or El Ninos in general generally we tend to get two to three of these dunkel flouter episodes um a winter what I didn't show here which would have complemented the last presentation quite nicely is um I kind of did a bit of a correlation with um other European countries and generally when we get a dunkel flouter type of event very often it's accompanied by dunkel flouter in a neighboring region which is makes it hard for us increasingly harder if you like and then yeah I mentioned at the start of the presentation um that we have this unprecedentedly early SSW event so it's just about to tilt into a reversal of the stratospheric polar vortex probably today what's this mean um I guess anybody that's involved in you know forecasting we'll probably have an idea about what SSWs can mean um generally they can tilt the odds towards colder weather there's a little bit of a lag four to six weeks usually but what's interesting about this one is it is unusually early and this is called a dinosaur graph by the way hence the title there so it does not definitely does not automatically guarantee cold weather down the line it's way more complicated than that you've got to have this kind of chain of events to set up but it does tilt the balance slightly towards that and interestingly you can definitely see that it's already been priced into gas prices whether it unfolds or not we have to wait and see how it plays out it looks like it's going to be what's called a displacement event rather than a total split of the polar vortex but it definitely kind of increases the risk of some colder weather in December if you like now pointing in the other direction if you look at some of the longer range forecast models on the left here everything is pointing towards a mild winter and then compounding that we've got quite warm sea surface temperatures around the UK which tend to kind of ameliorate any kind of cold weather that reaches our shores so to try to tie all these things together it's a really kind of low confidence scenario to try and forecast the outcome this winter I suppose what we would say is that regardless if it ends up being a mild winter you're always going to get some sub-seasonal periods of cold weather and these can be accompanied by dunkel flouty conditions and stressful events so who would be a who would be a meteorologist but it's an exceptionally low confidence winter compared to yeah the summer that we've just gone into which we could say with a very high degree confidence that it was going to be mild hot if you like and definitely the photo in the background there is not a forecast for a white christmas pardon my ignorance but what's ssw it's a sudden stratospheric warming event okay that's when the stratosphere rapidly warms and it can tend to act reverse the usual west to east flow of winds in the stratosphere and that can percolate down to the surface and that can open the door to colder weather patterns in europe or and the two words i've been hearing quite a lot over the past weeks is the polar vortex and if i understand it correctly it's not that you know it's the fact that it's weakening right isn't it that the polar vortex is weakening that could have an impact potentially and how is it does it link to el nino at all it is linked partially to el nino yes it's a kind of complicated flow of events but yeah i mean it has weakened and it's just about to reverse its usual direction probably today coincidentally and that could be an indication of colder weather four to six weeks down the line is that what you always say exactly it tilts the balance in that direction yeah but you could be reliant on some other factors at play it doesn't automatically guarantee it but it makes it more likely yes i'm fascinated by the you know i think it's excellent that the english language has taken on board that german word dunkelflaute but what what's the equivalent in summer when there's no wind and it's quite hot do we then say hitzeflaute or hitzeflaute yes so that's actually a term that's used exactly and this would be when you probably got high demand cooling demand and you've got low wind and then yeah i guess rapidly decreasing solar ramp in the evening this can be a cocktail of events that spikes prices yeah which we've seen certainly exactly yeah that's already um final question for me rob but so you did use the term or the word unprecedented a lot and we saw the summer of 2026 was was was very extreme um it what summer 2027 28 are we going to get even even hotter uh summers way too early to say i mean basically if you it's basically a normal distribution of temperatures right and every kind of 0.1 degree of global warming you're moving that normal distribution temperature curve a little bit to the right which basically increases those return periods for hot summers so you don't have to move it very far to tilt the tail of the distribution curve so i've completely kind of skirted your question i think it's it's yeah it's certainly very possible given the trends that we've seen in the last few years when i started my career as a meteorologist in 2004 i didn't think it'd be possible to see 40 degrees in the uk we have right and we almost reached it this summer as well so it wouldn't surprise me at all thanks rob any questions for rob there's one at the back here and then one i'll take the one in the front first and then i'll move to the back hi either mccrishan from north cadre and i wondered you said that there wasn't a correlation or you didn't see evidence for a correlation between dunkel flauta events and el nino but is there a correlation for dunkel flauta events in time as the years go on i just wondered from that bar plot yeah we did we did actually do this analysis on behalf of a client recently and we didn't see that there was any statistical trend in dunkel flauta events and time so no nothing that we could detect i didn't include it in this presentation but yeah we have done that analysis i think probably it's more of the impact of dunkel flauta event of events that are increasingly or the consequences of them given the volume of renewables that are on the system now the consequences are more but they're not occurring more frequently i mean provided the critical situations that you show in your presentation at different stages right but the one that happened in october seems to be particularly serious this is a question not necessarily related with climatology or whatever but is there a sort of penalty for having that sort of missing forecasting in wind generation for the asset owners as far as you're concerned because obviously the lack of planification and proper forecasting might have actually generated a blackout i don't know if you're aware of that or if you're aware of that there's probably people in this room that are much better qualified to answer that question than i am to be honest my understanding of it is that you've got what's notified to niso but then you've got what's used for trading right and there'd be more penalty on the trading side than what's notified to niso but i'm not such an expert here that i'd be i'd quite dare to answer that question there are probably many people who can answer that question in the break and final question from you thanks for a great talk rob the suffolk school on the 9th of october and the sudden drop in demand of those wind farms could you run any of your models to try and identify the clusters of wind farms particularly around gb but perhaps elsewhere that might be more vulnerable to those kinds of events and is there anything about that bit of the suffolk coast that means that it's more likely to happen there than scotland or anywhere else i don't think there's any particular meteorological reason where why those wind farms would be more vulnerable than any others to be honest i think what does make make the system more vulnerable is that you have these big clusters of offshore wind right and you can have a scenario where all of a sudden a lot of production drops off it could be yeah it could be a spike or it could drop off through one of these type of events or when there's a high wind speed shut down that that could be quite challenging for NISO to to deal with fantastic rob once again many thanks and here's here's a little present from us thank you thank you