Derecho

Weather-history It goes back to the correlation does not equate to causation argument. I can't see how a pattern several months beforehand can lead to something several months later at our latitude. In the tropics we get some monthly persistence with the likes of ENSO but that's as good as we have right now. One thing I have noticed since the 2010s is that when we get an Atlantic dominated period, it tends to be associated with high pressure over Spain rather then the Azores. This means rather then a mix of mild and chilly westerlies with a northerly toppler or a ridge never far away. Now it's just endless days of SW'ly winds with a pattern hard to shift once it sets in. I also think the west-east progression of low pressure systems is slowing down and patterns get stuck for longer. I wrote a similar post mortem for winter 2018/19 and I still have those same beliefs now. I do think at some point we will shift to a bunch of blocked winters but we are living in changing times.

Earthshine Interesting. From working with reanalysis models in the past, I know they are able to go far back in time because they assimilate little in the way of variables. The 20CR for example goes back to the early 1800s because it only assimilates SLP, SST and to an extent sea ice into a weather model. I'm guessing that running them in the level of detail to capture small scale events like local convection just isn't possible with the number of variables in earlier years and computing power available? So are other methods are required to capture this detail and still being worked upon?

Frigid The MJO hasn't been in a favourable phase since the first half of January and some blocking did follow. It's being used as a bizarre scapegoat.

Earthshine Seems like reliance on reanalysis models constrains the output of AI models then. Maybe tweak them to apply the laws of physics in areas where we have a deepening low only? Leave the rest to AI? The interesting thing is... given the AI performs just as good and potentially better then the physics driven models, there must be aspects of physics the AI model is incorporating that we don't understand yet?

It's an interesting response but the winter months of 2010 saw the complete opposite. Warm air often moving into western Africa and southern Spain with anomalous warmth in the SE North Atlantic. So what made 2023/24 so different?

Bristawl Si That logic wouldn't apply further into the past though. Between the 1740s and 1820s easterly winds were nearly as common as westerly winds during winter in this part of the world. If we are to see colder winters again though, changes in ocean circulation are required IMO to overcome the expansion of the Hadley Cell. 2009 and 2010 show it is still a possibility. jules216 Agreed. I think the February 2018 BFTE resulted in too much confidence in the belief that an SSW = UK cold. Yes it can split the vortex but the pattern has to fall favourably for the UK to benefit.

Nick F Have to disagree to some extent... the MJO composites correlated very well throughout the winter which is why I became more of a fan of them. For example in late November we saw the period of blocking that arguably ended up producing the best week of winter weather here. Phase 6 and 7 in late November. This was following on from a northerly which the MJO composite below fits very well in. Then through December we had a very unfavourable progression through the MJO as it went round to phase 6 and then a much hyped move into phase 7 broke down before christmas which meant we ended up writing the rest of December off. We then went through phase 1-3 at the end of December and into January. Our most blocked phase of the winter followed. Then we went through a strong phase 5 and 6 Phase 7 isn't favourable in February either as it still favours higher then average pressure over Iberia. So yeah I somewhat disagree with the interpretation of some of the composites but in the face of a changing climate you may ultimately be correct. You can find them here MJO.html WWW.AMERICANWX.COM Bear in mind I've been using the El Nino composites this year. Phases 1-3 are cold in January whilst phase 4-6 are more Atlantic dominated. The issue is the reliability of forecasting the MJO when it is in a less amplified phase or there is a long time difference. This caught out those ramping up a phase 7 in December and I pointed that out at the time there was low confidence of that occurring. There are flaws though. The sample sizes for some phases can be quite small and we don't know how climate change is affecting their response. When using multiple other factors I'm not sure what can be effectively concluded other then speculation given the large number of variables and diminishing analogues. I don't treat the MJO as gospel to me they are a bit of speculatory fun but I believe they are a useful tool for determining whether we have scope for coming out of a SW'ly rut. Given that has been the case so much this winter, they've been a straw to clutch in what has been a very poor 3 months. A colder winter and I probably wouldn't have paid as much attention to them.

Cheshire Freeze I hope not given the amount of rainfall we've seen in recent weeks! I suspect you are thinking of 2007 as an analogue... lol. Other similar years would be 2010 and 2016. The former was quite poor but not to the same level as 2007/2012. 2016 I thought was pretty reasonable. I wouldn't mind a summer like that with more in the way of big thundery breakdowns. The only matter of interest for me this winter is any potential SSW and the prospect of a phase 8 MJO but the GFS and EC differ from day 1! With the EC keeping things in the COD. Can't rule out more cold and snow yet but it'll be difficult to get something notable I reckon. Hopefully next winter eh coldies?

Cheshire Freeze Problem is even with a minor decrease in rainfall, the risk of drought is enhanced by the fact it is much warmer then decades gone by. I agree with @Daniel* though in that precip trends are very uncertain in the UK overall. We don't know what atmospheric circulation will do in the future. Due to the persistence of anticyclonic conditions this winter I do seriously worry about Spain this summer though Spring can also be wet so it is still early days. I think some cloud seeding in Spain is what we need , end their drought, boost the snow in the Alps and improve our winter prospects. If only it was that simple ha. Or for ocean circulation in the North Atlantic to relax like it did in 2009-10. I suspect the warm SSTs in the subtropical Atlantic are giving any long fetch SW'lies a bit of extra warmth in recent months. This winter seems to oscillate from briefly chilly to exceptionally mild. Looking long range a quick switch to a La Nina appears to be on the cards. That and the very warm tropical Atlantic which should persist through the year means the start of next winter may well be promising with ridging in the North Atlantic towards Greenland, especially if the PDO is strongly negative. Mark my words, I think the Atlantic hurricane season this year will be a really bad one unfortunately. As you can tell I have given up on this winter...

I think the EC and UKMO have a minor advantage but that's it. There are occasions when the GFS comes out on top such as this weekend but the differences are generally quite small. My way of thinking these days is to look at all the model output. Then answer the question 'which model produces the mildest weather?' and I have my answer right there for the correct scenario We do have our background signals but they are not favourable for now so I think the milder solutions have it for next week and probably for much of February. The global temperature anomaly chart sums up this winter perfectly. Cold nearby but we end up on the mild side of a battleground and endure the unsettled conditions that often come with it. You can't help but wonder, had global temps been colder, would that cold pooling over Scandi have been more expansive and hence filtered into our daily weather a lot more?

It's worth adding about the 1779 record, there is some level of scepticism about how warm the CET mean is for that month. I read a paper that suggested it was a fair bit cooler then 7.9C but it was quite a while ago so I can't remember the paper in question unfortunately!

I remember this day as it all happened still as I was still working as a meteorologist at this point. It is a day that will stick in my mind for the rest of my life, seeing the obs come in was jaw dropping. Anyhow I remember the hottest temps in the south were always going to be a bit earlier in the afternoon so Cambridgeshire / Lincs was the most favoured area to beat the record as the hottest air cleared here a little bit later. During that Tuesday morning, there was some high cloud over Lincolnshire / South Yorkshire and if that hadn't been there I think an official station would have recorded 41C during that afternoon. I also remember the very hottest area in the model forecasts (somewhere near Worksop) had a lack of nearby official stations so somewhere in that area could have seen a max temp above that seen in Coningsby. I remember the UKV had the highest maxes with some going for 42C but fell back into line 3 days beforehand (see below). The EC underplayed the maxes for quite a while even when bias corrected (EC undercooks maxes by 2-3C in the summer generally). GFS overcooked the maxes which is what the BBC was based upon.

Jason M It'll be interesting for sure though I think models are bullish on a return to a La Nina this year. El Nino conditions should quickly decline but its footprint will be slower to filter out of the atmosphere. Sods law means we will probably see a cold April like 1998 or 2016.... Summer forecasting as we all know is even more chaotic but it wouldn't be surprising if we saw another 35C again. Aside from that I reckon we'll get an exceptionally active hurricane season (perhaps record breaking with Tropical SSTs being so high) and a better shot at northern blocking early on next winter.

WeatherArc Yup, out of all the charts today this frame from the EC at T240 really caught my eye though probably wouldn't verify of course. If that was over the UK in May the uppers would be reasonably impressive in years gone by but to see a plume that strong in mid-February really is quite something.

Very true!, It did feel like spring for a brief time here on Saturday but otherwise it was largely cloudy and drizzly at times. I'm looking forward to the blue skies in the strengthening sun. That to me is when it feels like it has really arrived! Knowing our luck though we will get a cold sleety March.

If that high nudges further east we can get a nice early taste of Spring! Given that early February was supposed to be dry with winds of an easterly component in the MetO long range forecasts, you wouldn't bet against the mildest solution possible verifying.

It seems the only SSWs to affect the UK this winter are Sudden South Westerlies... A struggle to get into MJO phase 8, record high February global temperatures and a lack of northern blocking have worked against us in what has been a stinker of a winter. I think I'll only get interested again if we see another beast from the east. The EC does go into phase 8 but it takes too long. We'd only feel the effects in early March. By that point I will be bullishly looking for the first 21C rather then having increased heating bills when the best we'd probably get by March is most likely snow showers with accumulations quickly melting by day (unless we got something really notable). Roll on Spring! I've had enough of winter 2023/24!

Decent output tonight. Though I'd like the MJO to go into phase 8, phase 7 isn't the be all end all. That phase is just less significant for our part of the world and others drivers come into play. It's a very messy picture next week but we seem to be creeping towards a snowy situation with unremarkable synoptics instead of a huge block with loads of cold nearby. I'm therefore quite relaxed about what happens next week, though as others have mentioned these events seem to trend south as time gets nearer. My journey north from Nottingham on Thursday might be an interesting one.

northwestsnow MJO phase 7 isn't the best composite in our part of the world. Its signal is rather weak besides the ESB low and higher pressure over the Arctic. Phase 8 is what we want to see. The GFS this morning is hinting at it. EC 00z wasn't on board with the amplitude waning in phase 7.

With the phase 7 transition coming right at the start of the month, I was unsure what kind of synoptics we'd get. Clearly we are leaning towards this composite for a brief time rather then the much more favourable January composite. Then a loop into phase 6 if the EC is to be believed (the GFS lingers in phase 7). The output does bear some similarities to what we see below: More so with the phase 6 composite on the right but note the phase 7 composite has the Iberian positive GPH anomaly along with to the NW. Again we will struggle to get notable cold until that goes. Phase 6 doesn't favour an Iberian high anomaly though that may be brief if the below plot is anything to go by. Any progression into phase 8 of the MJO looks highly uncertain and painfully slow if it does. A shame because that composite is highly favourable towards cold. Though we seem to be hitting a brick wall in phase 7. The EC does make it towards the end of February but we probably wouldn't see a response until early March. Sadly I'm running out of patience with this winter. It hasn't been a good one. February may still throw up a surprise but I'm highly doubtful now. The soil temperatures from the mild spell and the increasing solar radiation as we go through this month means it'll be tricky for snow to stick even in great synoptics unless we get exceptional cold. I'd rather have some nice weather and save on heating bills unless we get a beast from the east. Come on models, show us some improvement in the next few days otherwise I'll be going into summer mode.

@Relativistic Indeed it does but given the complex nature of the current models, there is a black box element to them anyway. Either way I think there will be big advances in forecasting over the next decade. For example, how much better will 40 years worth of observations / analysis data going into a model compare to theoretical assumptions that are currently used over the Greenland region?

Very interesting article this. I'll put the key pieces of text in below for those who want a more simplified outline: Just a moment... WWW.SCIENCE.ORG Today, the 6-day forecast is about as good as the 3-day forecast from 30 years ago. It also comes with a cost: billions of dollars’ worth of energy-hungry supercomputers that must run 24/7 just to produce a few forecasts a day. In mere minutes on cheap desktop computers, trained AI systems can now make 10-day forecasts that are as good as the best traditional models—and in some cases even better. The algorithms could enable more frequent forecasts and free up computing resources. Traditional weather models start by feeding a snapshot of current conditions, based on observations from satellites, weather stations, and buoys, into a gridlike computer model that divides the atmosphere into millions of boxes. The snapshot is run forward in time by applying the physical laws of fluid dynamics to each box. The new AI models skip the expense of solving equations in favor of “deep learning.” They identify patterns in the way the atmosphere naturally evolves, after training on 40 years of ECMWF “reanalysis” data—a combination of observations and short-term model forecasts that represents modelers’ best and most complete picture of past weather. When fed a starting snapshot of the atmosphere based on the same combination of observations and modeling, GraphCast can outperform the ECMWF forecast out to 10 days on 90% of its verification targets, including hurricane tracks and extreme temperatures. To improve further, the AI models could be weaned off the reanalysis data, which carry the biases of traditional models. Instead, they could learn directly from the petabytes of raw observation data held by weather agencies, Keisler says. Google’s short-term weather model already does so, training itself on data from weather stations, radar, and satellites. However, note the final paragraph: Adoption might be slowed by unease about the black-box nature of the AI: Researchers often can’t say how such systems reach their conclusions. So instead of relying on physical theory to produce weather models, AI is taking past reanalysis models to train data. This can be run several times a day and is currently comparable to EC model output and in some cases better. When we start assimilating even more observations from the past we should see even more improvements... but how good can AI forecasting become?

@Metwatch Indeed but can we achieve that with an airmass as hot as the one we saw in July 2022? I think yes but it'd require all the jigsaw pieces together. I think high summer will see more of these transient but intense plumes with the regularity of these reaching 40C somewhere increasing. One year though we will see something off the scale.

@WYorksWeather Could well have done, the forcing behind the build of any pressure to our east would have to overwhelm any Atlantic push. It's definitely possible. I find the long heatwaves are often the most uncomfortable. For me August 2020s heatwave felt a lot worse then July 2022 because he had 32C+ for about 6 consecutive days I think. I like to make educated guesses into how hot things might get in the future but I think we will continue to see surprises.

@raz.org.rain The thing is, prolonged phases above 30C are more likely to be prolonged days of anticyclonic weather with temps of 31-33C typical. The hottest days are often plumey now and this heat intensifies the temperature gradient between the oceans and continent. That in turn often fires up the Atlantic which explains the transient nature of the most intense heatwaves. However at some point I do think there could be a plume that does bring heat to the UK and gets stuck. That would be exceptional and also very unpleasant. In a warming climate I would still expect that synoptic to be very difficult to achieve.