Man With Beard

That's amazing Roger, thank you. Who on earth would have thought a near record June to September period could include such a dodgy 7 week period in July/August, potentially beating 1976 and 2018. Statistically, then, one could argue this was one of the best summers ever if September were considered a "bonus" summer month!

Thanks for the reply! Oops, sorry I meant the 4 month period June, July, August, September (my fault, terrible wording, and I'm supposed to be a university linguist by profession, good grief )

Has anyone looked into this? What is the warmest June-September CET combination on record? Seeing that June was the warmest for 200 years and September might be the warmest ever, surely 2023 might come close to the top of such a table?

I suspect September will be remembered, as it was clearly the best September heatwave in recorded history by some margin. However, I'm not so sure about June - it was consistently very warm in the CET zone, but it wasn't always warm in some spots like the east coast, and it didn't have the qualities of a classic heatwave, which I would call a long period of consecutive days above 30C with the peak around 34C or more - all of the heatwaves considered historic qualify with this criteria (e.g. 1976, 1995, 2003, 2006, 2018). 1983 was similar - consistently hot, but no signature heatwave in its own right.

Is it correct that only 4 months in the whole of recorded history have had a higher CET than this? I realise this is only half a month, but this stat does put the early September heatwave into perspective.

I'm thinking it will be a 50/50 call as to whether the CET record is broken. I guess we'll be around the 19C mark at halfway, meaning we'll need to average 15C for the rest of the month to get the record. That won't be exceptionally easy - it's certainly doable, but with an Atlantic driven week coming up next week, I can't see the CET for the period 18th-25th being much in excess of this figure, and it might potentially be below it.

Sorry just seen this ... I double checked on weatheronline and it was actually 5 days above 30C in June (June 10-13 and June 25) - a lot of 28s and 29s though!

I think it was 6 in June, 1 in July and (likely to be) 7 this month. *I was thinking of summer as to mean the whole of the "warm" part of 2023

By tomorrow, I make it there will have been 14 30C+ days this summer. That's quite a decent number and would compare favourably with a number of summers considered above average.

Add the ECM to the list for Saturday - raw max temps 32C, which is 2C higher than it has for today

Sunday hotter than Saturday on most models

Don't know where to put this...

I think this spell will break a lot of records. It looks very possible we will get 7 consecutive days above 30C in this spell - I'm going to gamble there's never even been 4 consecutive days above 30C in any previously recorded September. CET likely to be above 20C for the first 1/3 of September, again, surely never happened even in 1911. I note the ECM is also returning to its idea of increasing the heat yet more for the weekend - raw 31C on Sunday's chart, a degree higher than tomorrow's chart - this puts an actual 33C/34C in the frame going by typical adjustments - and every now and then the ECM has been known to be 4C too low in a plume, so even a very tiny chance of 35C??

Of course, claiming the September record won't be guaranteed even if we're 20C by mid month. The rest of the month would still need to average 14C, which it seems to me would still be slightly above average for the second half of September. Models increasingly converging on an end-point for the heatwave between Sunday and Monday, maybe Tuesday, so probably not 20C by mid month, but likely to be over 20C around the 12th.

GFS 12Z: Uppers 19C or more over the SE for 84 consecutive hours between Monday night and Friday morning. UKMO similar. Very rare even at the height of summer. ARPEGE gets to 34C by Wednesday - close to the September record - looks a bit high compared to other models but surely not out of the question. An outstanding heatwave regardless of the time of year.

I've seen GFS analysis today suggesting the CET might be around 20C by mid month - unlikely but wow!

I've been thinking that for years! Nice to find out the pros thought the same - although in the last 15 months GFS has switched to being slightly overcooked if anything?

To compare, ECM (raw) not a SE only event at all, this is Wednesday, it shows 24/25C in many northern areas of England and high 20s in Midlands - likely to be slightly under baked considering ECM's track record

Not sure if the ECM 00Z ensembles were covered at any length, but very solid support for a heatwave in the SE from 4th September to 9th September, with no member showing uppers below 12C during this period and the majority not below 16C. If this period of heat had arrived even just 10 days ago, it would have led to several days in the 90Fs. We lose the heat potential so quickly between mid August and mid September.

Hottest day of the year ... in September?

UKMO still puts us much closer to the hot stuff than the other models. 16C uppers for London by Thursday on the 12Z. Support from GEFS ensembles at about 15%, ECM 06Z ensembles more like 10%. Down from 20C uppers for a short period on its 00Z run - which had practically zero support from other ensembles. Is this another phantom chase from the UKMO? Or is it the genius in the pack?

There has certainly been more of a tendency for models to correct towards the east than usual in the past two months. It seems to happen to a greater or lesser degree over a period of time, and I feel I've seen this a lot in recent Augusts. In this occasion, it may be to our benefit as you mention, with the high pressure now toppling over us during the Bank Holiday instead of holding north-west. In the nearer timeframe, the ECM has found a little extra push from the continent this Thursday and now suggests a raw 28C in the SE corner - might we get one final 30C for the year after all? Still an outside bet but getting closer.

The ECM and UKMO seem united tonight in prolonging the warmth (heat?) for the south east corner well into the week, the ECM looks likely to mean mid to high twenties even into Friday now. Acceptably warm through most of England, less good north and west. The cool down for the bank holiday potentially being watered down, too.

ECM ensembles tonight ... jumpers at the ready for the end of August?

Time for an uncomfortable discussion: Is the UKMO still fit to be considered one of the "big three"? This used to be the "reliable" model. The one that didn't go with every whim, the one that was maybe over-conservative at times. We used to say "don't believe it until the UKMO is on board". Not anymore I feel. Today was one example. The difference between the 00Z and 12Z was pretty huge, considering it is a D7 maximum model. Way beyond the normal swings you'd expect at such a short timespan. And this, I feel, is becoming regular for the model. I imagine the global verification stats suggest it is still "ok", but my gut is telling me ... it simply isn't worth the time that following the ECM or GFS is anymore. Thoughts?