Spring 2024

[damianslaw]

 Frigid The last 3 big el nino winters were followed by average to cold Marches and Aprils, 1983, 1998 and 2016..

[WYorksWeather]

 Shillitocettwo The March and April CETs are probably just about within the realms of possibility but I have to say that in the current climate a May like that is impossible. The March 30C isn't possible either. The April couple of days at 30C is possible, just about at the tail end of the month.

Depending on how things go though, that might be on the cool side for spring 2124 though.

[Greyhound81]

 WYorksWeather

Do you think 35C is possible in May?

[raz.org.rain]

 Greyhound81 what do you reckon @CryoraptorA303

[LetItSnow!]

I think we’ll probably have a warm and wet spring overall, but I think one month will buck the trend, either in terms of being much colder and much drier. Perhaps a spring like 1998, which had a very mild and I think wet March, near normal April that was very wet, but then a warm and very dry May. It may not pan out exactly in that order though; perhaps we’ll have a dry and warm April and an unsettled May. Either way I have a guess that the spring will be like 1998 or 2007, both similar springs just in a different order. I think we’ll probably see a potent northerly at some point that we would have bitten an arm off to have had in February. 

In 1998 there was a static temperature pattern between February and April with 7.2, 7.9 and 7.7 being the CET’s off the top of my head. Wouldn’t be surprised if something like that occurred. 

Edited February 20 by LetItSnow!

[CryoraptorA303]

 Greyhound81  raz.org.rain I've done the maths in another thread, I'm not sure which one. Exceeding the current 32.8°C record is possible at any point in May and even in late April. 35°C should be possible from around the 20th, and right in the last few days 36°C would be possible.

I should also add that 30°C is just about possible right at the end of March. As Easter Sunday is on March 31st this year, this will be the best opportunity to see the Easter record broken on a March Easter for a long time. Thus 30°C is possible anytime in April although it becomes significantly more probable as the month goes on. As mentioned, even 33°C is possible at the tail end of April.

This calculation is based on the maximum anomaly seen during Coningsby, which was +17.6°C at Nottingham, applied to the hottest stations of the country.

This may seem like an unacceptable jump from the historical extremes, however it must be kept in mind that the hottest historical temperature on record was 36.7°C and Coningsby exceeded this by 3.6°C. If you want to go even further, that record was made in early August and early August has always seen the very highest temps in this country, so there is the argument to be made that Coningsby would see an even higher temperature if it were to have happened in the first 10 days of August 2022. The historical record for July was 35.9°C in 1976, and Coningsby exceeded this by 4.4°C.

Just for fun, here is what every month could expect to see were it to experience a Coningsby-level heatwave:

January: 22.7°C (18.3°C historical record)
February: 23.4°C (19.0°C historical record)
March: 30.0°C (25.6°C historical record)
April: 32.5°C (28.1°C historical record)
May: 37.2°C (32.8°C historical record)
June: 39.9°C (35.5°C historical record)
July: 40.3°C (35.9°C historical record)
August: 41.1°C (36.7°C historical record)
September: 40.0°C (35.6°C historical record)
October: 33.8°C (29.4°C historical record)
November: 25.5°C (21.1°C historical record)
December: 22.7°C (18.3°C historical record)

All records used were set before 1990, which is generally considered to be the boundary between historical and climate change times.

Records in italics are very old records that may not be exact. 29.4°C on April 16th 1949, 35.6°C on June 29th 1957 and 35.6°C on June 28th 1976 have been omitted due to questionable accuracy and are not counted. Instead, 28.1°C on April 21st 1893 and 35.5°C on June 27th 1976 are used. 21.1°C for November and 35.6°C for September are unlikely to be exact, with the latter likely being somewhat of an overshoot and the former being a rounded Fahrenheit reading that was seen at eight different stations, so we'll never know the exact number.

Of course, the hottest air was over us at night during Coningsby, so we know that 40.3°C isn't the ceiling for July and 2-3°C higher can be realistically achieved with better timing and/or a more intense heatwave, so hypothetically these extremes can go even higher in the current climate.

 

[baddie]

Lets face it, we arent getting an SSW. It would probably fail because it has been a mild and wet winter, if it does, I can see something like a March 2021 style heatwave around the 23rd-26th, with 25c seen, then straight to cold and snowy by the Easter Weekend (Resembling April 2021)

[WYorksWeather]

 Greyhound81 I think 35C is just barely possible at the end of May.

 CryoraptorA303 Those numbers strike me as just a little on the high side. I'll admit that part of that is that I don't really want to believe that they're possible.

I feel that were I manually adjusting your projected temperatures, I'd add a degree or so in high summer, but subtract about a degree or so at most other times of the year, maybe 2C from November to March when solar heating is very poor (November to January) and/or seas are very cold (January to March).

The final point, and probably the most debatable, is what do we mean by possible?  My hunch is that Coningsby is still a fairly rare event in the current climate. But if we look at some historic ones, my vote for the most insane heat event in UK history of our monthly records is the 35.6C at Bawtry on 2nd September 1906. 

The anomaly over the 1961-1990 CET average max (not that dissimilar to 1906) is +18C, which should be rounded up to about +20C given how far north Bawtry is. You then add the rough 1C increase in extremes per decade over the last three or four decades, so the Bawtry analogue today should be the peak July or August 1961-1990 CET maxima plus about 23C or 24C, which should give 43-45C depending on the exact assumptions. However, given how well that event has survived various attempts, it might be something like a 1 in 1000 year event in terms of synoptics.

That's where the word possible does a lot of heavy lifting. Your figures are a bit high if you're thinking about records we can actually expect to set in the next few years, since a Coningsby type event is probably a 1 in 10 or 1 in 20 year event, so it would take a long time to replace all the monthly records. But an event like Bawtry is the other extreme - it's the answer to: 'what temperature is achievable if we completely load the dice?'

 

[Rain All Night]

 WYorksWeather That looks nice...

[Methuselah]

Oh no, not another one:       

[LetItSnow!]

 Rain All Night Just flip the pattern around for a nice cool, crisp, bright northerly and that would be much nicer. 

[CryoraptorA303]

 WYorksWeather Well, this is what is possible via the Coningsby anomaly in each month. I'm not necessarily expecting us to see these temperatures in the next few years.

September 1906 is what happens when you have Coningsby but the hottest air is over you in the middle of the day.

If we compare September 2nd 1906:

With July 19th 2022:

You can see that they are overall quite similar. Such a heatwave today would likely see 37-38°C and 42-43°C in mid-July to early August as Coningsby would have seen had the hottest air coincided with daytime.

Note that the quoted values aren't the highest possible, these are just if the exact scenario of Coningsby played out in the other months. As stated Coningsby was capable of sending us to 42-43°C. I've also calculated that a repeat of August 2003 today would see a maximum of 41.6-41.9°C, based on the acceleration of extremes since then.

An exact Coningsby repeat is still going to be relatively uncommon if not rare, however we don't need a Coningsby setup to exceed 40.3°C. As mentioned, the August 2003 repeat would in all likelihood quite severely exceed it, even though the exact synoptics wouldn't be as severe, purely due to the timing being more precise. With each passing year it will take a less and less severe synoptic to surpass 40.3°C, only requiring that the timing isn't as sloppy as Coningsby. When you think about it, it's timing was very poor, with it being only mid-July and not in the hottest time of year with the ideal insolation:ocean/airmass temperature ratio, and the hottest air by 2°C being above us at night. August 1990 was a bit inexact, but it did at least strike at the most ideal insolation:ocean/airmass temperature ratio of the year, and it was only 1990 and climate change had not worked it's magic just yet. August 2003 was almost perfectly timed with the very hottest air moving over in the middle of the day, right at the end of the ideal ratio of the year. A repeat of that this summer would surpass Coningsby by some margin.

In the same vein, we don't need as extreme of this margin to beat the historic records. A Coningsby in April would give us approximately 32.5°C but a considerably less severe heatwave could still give us 30.5°C right at the end of April and still break the record. A Coningsby in May would give us approx. 37.2°C but a significantly less severe heatwave in the last week could still give us 34°C, and so on. In contrast an even more severe or better-timed event can give us even higher temperatures, but outside of the ideal ratio it's diminishing returns.

Of course I don't expect every month, or any month to reach these figures anytime soon, although I suspect at least one month this year will be having us doing a double take.

These extremes have happened elsewhere as well; a few years back, I think it was either Montreal or Toronto (or perhaps both) that had an extreme May heatwave and recorded one of their hottest days ever, even hotter than their hottest June day. Of course they are considerably different climates to us, but such feats are clearly possible.

Edited February 20 by CryoraptorA303

[WYorksWeather]

 CryoraptorA303 Yeah I wasn't really attacking the method in that sense as it is just using one measurement as a proxy (namely, the Coningsby anomaly). I was just pointing out that the approach might be less applicable as you move away from the peak high summer period.

The reason I flag the Bawtry heatwave though is because it also got the focus wrong. Look at the position of the cut-off low. It's dragging the hottest of the air over northern and central England. So when you adjust for that as well, that's where the other couple of degrees comes from, which pushes it above Coningsby, since it achieved a very long-lasting record, which has withstood numerous attempts including several in the christmas pudding (most notably of course September 2023), without even focusing the heat over the right part of the country!

At the right time of the year, and with better positioning, it would be our answer to the 2021 British Columbia heatwave. The sort of event that doesn't just break records, but absolutely demolishes them. I don't actually think we've had any events like Bawtry 1906 yet in the christmas pudding since 1990, in any month. There just hasn't been long enough for an event to achieve the right combination of timing (coming right at the start of the month in question), background (a long, hot, and dry summer beforehand) and the synoptics.

In terms of severity, I think it's something that's difficult to assess. I'm very wary of any attempts to guess how rare an event might be in the current climate. All we can probably say is that it's doubtful that Coningsby is very common, say one every five years or more, since we'd have seen it several times already. It is also doubtful that it is exceptionally rare, because of the number of times we've exceeded 38C in recent years, and the fact that it didn't absolutely destroy previous records. If I had to guess, I would say that if we froze CC today, I'd expect Coningsby to recur maybe once every 20-30 years. But of course that's not what is happening, so it's a bit of a moot point.

Anyway, to turn my attention back to spring and the topic of this thread. As I've been saying, what we need to see first of all is a dry spell. All else is secondary at this point. I would prefer warm and dry, but cold and dry is preferable to any further heavy rain. No more!

[Metwatch]

For England (ignoring Foehn for the winter months), my estimates on the highest possible temperatures in the next few years given current climate, rounded to the nearest degree:

January: 18°C
February: 22°C
March: 27°C
April: 31°C
May: 34°C
June: 37°C
July: 40°C
August: 39°C
September: 36°C
October: 30°C
November: 22°C
December: 18°C

I can't see England reaching 20c during December and January, simply not enough solar heating, and would take a lot more warming to get there and there would still not be enough solar heating. 30C for March it's virtually impossible as well at the moment (even Bordeaux hasn't seen 28C in March yet same with many areas in central France not seeing much above 26C) unless the world warms by another few degrees. April and October I think could see 30C in the next 10-20 years. August 40C, that might be possible in a few decades time, but not at the moment, even with the 40c in July, that I can't being repeated for the next 5-10 years at least (hopefully not a jinx and we don't see it this year).

@Rain All Night nah not in the next 5-10 years at least, I think we could get very close in one of the years this decade, maybe 39.0-39.5C though. 36.4C in August 2020 seemed to be the highest we've gotten in recent Augusts, can't see the jump all way to 40C from that.

Edited February 20 by Metwatch

[Rain All Night]

 Metwatch Do you not think 40C would be possible now in early August?

[Rain All Night]

 Metwatch And why is that, I'm assuming the reduced solar input compared to mid-July?

[Arctic Hare]

1 hour ago, WYorksWeather said:

a very long-lasting record, which has withstood numerous attempts including several in the christmas pudding (most notably of course September 2023)

Although of course records are only very long-lasting until one day they aren't. I suspect it will go before 2040, and I may be being a bit conservative with that. The 29.9 °C October record may be a trickier one given that happened on the 1st in an exceptional spell... but then why am I more doubtful about that than about the September record going? I'm not sure I could give a sensible answer that doesn't rely at least a little bit on emotion and hunches.

(Not up to me of course, but that "christmas pudding" auto-replace the site does really has long outlived its usefulness now, surely?)
 

 

[raz.org.rain]

 Arctic Hare a good rule of thumb is that such records can be expected to fall within a much shorter timeframe as CC continues to get worse. So while in the past, such a record might be held for decades, in the near future a record high could be beat within a few years of being set. 

[Greyhound81]

 CryoraptorA303 Intersting that you have March 4.5C higher than November, when March is usually the colder month of the two.

Edited February 21 by Greyhound81

[Dark Horse]

 raz.org.rain like the recent heat records then with it set in 2019 at Cambridge and then it got broken again in 2022 in Coningsby. I think the trend is clear because we're not beating cold records are we? It's getting extremely difficult to get an average CET month lately never mind a below average one.

[East Lancs Rain]

Only 9 days to go until meteorological spring but there’s no sign of anything springlike yet.. CFS Model is forecasting a delayed spring with northwesterlies in March.

[cheese]

 CryoraptorA303 August is definitely the next month to keep an eye on for 40C+ temperatures... maybe August will even reclaim its title as having the hottest temperature on record in the next few years.

Edited February 21 by cheese

[SunSean]

 Metwatch always find it odd how May struggles to reach the 30c mark. We haven't touched it since 2005 in my area!

[kold weather]

I actually think the July 22 heatwave was more exceptional than some are letting on, its utterly smashed the record hottest CET day, beat out the previous record comfortably as well. As other have said if it came along a touch later we'd have probably gotten a little higher. Then again you could argue we would not have had such an absurdly warm night time base to work from so the end result may have been the same.

I think what made the July heatwave particularly exceptional was it was a 2 day blast, it easy to forget that the day before was a near record breaker in itself. I think it sits either at no.2 or no.3 in terms if CET days.

The setup is nearly so loaded and progressive to get the heat in that to have it last 36hrs at its peak is what's real impressive to me.

@Dark Horsesince Jan 2015 we've had 1 cold record and 63 heat records.

To say the gun is loaded to heat is an understatement.

Edited February 21 by kold weather

[CryoraptorA303]

 Greyhound81 On average it has lower minima, but overall slightly higher maxima, and March will always have higher extremes. It may have some of the coldest water and airmass temps of the year (hence why March snow is generally more likely than November), but the Sun is much, much stronger than in November, by March 31st its as strong as early September.

 cheese I would bet on it even. I think these maxima all falling in July is a bit of a fluke; this is the longest period we've gone without an August annual maximum since at least 1875. It's probably a consequence of July occupying the most annual maximum likelihood at the same time we have a rapidly increasing rate of extremes, combined with the fact that we've had an awful lot of La Nina in the past 15 years and these tend to have more frontended seasons. And of course, some of it is just luck; August 2020 got scammed out of the fifth hottest day by it falling on July 31st. Literally 24 hours later and August would've had that one.

August historically had all the very hottest days, even 1976 didn't reach them and 1990 and 2003 were both in August.

 kold weather Actually, if Coningsby happened in early August then chances are we'd see an even higher daily minima as the oceans would be moderating even less and the airmass would've been even warmer. The onshore sea breeze that the strong southerly wind brought would've been considerably weaker by early August and the epicentre would be further south where the Sun is just a little stronger. Combined with the even warmer airmass and the ground having 2-3 weeks to dry out even more, we would've seen the London area, Surrey and East Anglia reaching 41°C and instead of Coningsby, we'd be calling it Charlwood, Cavendish or Cambridge in all likelihood, or maybe East Malling, Wisley or Writtle.

As for rarity, Coningsby is something that is still uncommon to say the least, but it was sloppy. The uppers we saw at night would've been enough to see 42-43°C in the daytime. In theory a less severe heatwave that is better timed like an August 2003 repeat can still exceed Coningsby without being overall as severe.