kold weather

18z GFS and Icon are still somehow on the cooler side with 36/37c respectively but that is an increase in earlier runs.ad to call 36c on the cooler side...but it is!

I can see the tops flashing away from here, providing they keep going I've got a great south facing vantage point over the next hour or so.

The discrete cells to the east are going to likely be very intense indeed. The stuff further west maybe not quite but it's going to likely form into a decent strength MCS over the next few hours.

Just something interesting to note, the ARPEGE ensemble members that develop cells further east tonight are also the runs that don't go as high on Wednesday or Thursday. Given we have decent cells developing quite far east just something to think about.

Both the GFS and the ICON have been both too low and far too restricted in the region of highest temperatures today, plenty of 32c reports and some 33c as well in a fairly stretched out area, the ICON was almost 3C too low across the board, and the GFS varied too low between 1-2C. ARPEGE was a hair too high but far closer than the GFS/ICON as it had predicted a widespread area of 32-34c, far better forecast than the GFS which restricted 30C+ to the M25.

Amazing image PM, rare you see 39c modelled it has to be said! Also one of the most widespread areas of 36-37c I've ever seen, truely a large scale heatwave.

Upto 32c here, probably the peak as well. Very hot indeed, though obviously not unheard of. Thursday however is locally closer to 35-36c, which would be very rare indeed. Models still seem pretty keen on 37-38c being the rough max. Thursday is going to be mega interesting to watch.

Yes 40c is possible, more likely a rounded up if it's going to happen, but perhaps more noteable is the large area at 34-36c. Anyway ARPEGE has done very well today, pretty close to spot on from what we can tell from the 3pm temperatures. Bodes well for a record attempt.

Looks like Thursdays 850hpa temps will be a solid 1-2c higher than they were on 10th August 2003, 20+ looks more widespread as well. So that again would give a ballpark Max of 39-40c in theory.

huge area of 37c+ temperatures there, plenty of scope for 39-40c if such a chart came off in the real hotspots.

Just for reference, max temperature was 30.5 yesterday, so a touch higher than any model expected for yesterday, though ARPEGE was indeed in the right ballpark for most. A cooler start today but now we really start to advect in hotter air and it's much sunnier generally in the SE. Met office expect 34, hard to disagree.

I will go for 39.1c at Heathrow. Yesterday nax temperature was 30.5c by the way

It cools down quite considerable amount to average for the weekend (indeed its a AWFUL weekend for E.A, with 1 inch of rain and temperatures of about 16-19C) but once that eases off and we get back into the weak ridging, things warm up nicely again to between 24-28c, so more than good enough and probably wouldn't be going anywhere fast either. Decent area of 32C+ on the ECM this run for Thursday, even if the peak heat is a little more stunted.

For sure, as Man with Beard correctly said, we are still well within the ballpark of both the July and all time record even if the ECM 12z is correct with the cooler temperatures overnight, as there are several well placed stations that could pick up the heat on that run better than if it was shunted east of London. Personally, I'm happy with the 12z ECM, in my area it actually is an upgrade in temperatures, though still doesn't appear quite as high as it was on the June heatwave (as we were not far from the main heat zone on that one). Quite a bland weekend temperature wise, cooler than the 12z GFS for example (temps seem widely 21-23C, so fairly average). Indeed VERY wet weekend for East Anglia on this run, 25-30mm of rain!

I have to admit its lower than what I'd expect with the profile as it is, however we saw a few weeks ago that the surface conditions don't always reflect the crazy set-up above it. Besides, for perspective, the 36c the ECM has has only been beaten a handful of times, its still a really exceptional shot of heat by any standards.

Yeah that seems the route cause, though its very much the night time temps that are the limiter as explained above (IE, we get the same level of increase in temperatures, just we start from a 1-2c lower base to start with, so we end up 1-2c lower).

ECM 12z has a max of 35-36c in London on Thursday, still very hot for sure but the much cooler overnight temperatures compared to the 00z is what stops any record attempt. Also worth noting the other difference is the heat concentration looks a little more widespread than on the 00z. knocker - I think the main difference is purely those night time temps, the 00z had 66f at 2am over London (which is about the lowest it goes, it rises from that point) whilst the 12z ECM has 62-63f. That 3-4f difference makes the difference from the 96f of this run, and the 100f of the previous run. Simple as that from what I can see.

Still hot, but noticeably cooler than some of the other models out there tonught. Seems to stem from the overnight mins dropping away quite nicely. For example, London down to 17c by 1am... Looks somewhat too low to me though based on what all the other models are showing...almost certainly the coolest of the main models for sure tonight...

Whilst I've not seen past 03z on Thursday on the ECM, much cooler night than was forecasted by the 00z (by a solid 2-3c), so the heat may not be quite as extreme on the 12z ECM at the surface for Thursday day, but we will see.

I think the 6th did break the record but was broken more convincingly on the 10th. The 10th was indeed a pretty short notice affair as we pulled back in severe heat from France briefly.

What is quite interesting is the models have quite a spread of where those hottest temperatures are going to be, some are nearer S.Midlands, whilst some are London region. Regardless all models have expanded the area with 30-35c quite considerably.

In theory the June plume had temperatures that would have roughly been in the 41-42c mark at the surface had we not had that stupid ENE undercut. Anyway models still slightly adjusting westwards brining the more severe heat westwards with it. Increasingly becoming a more widespread heatwave. Most models still showing a Max of 35-38c, and most have widespread 30c now as well.

ARPEGE does look perhaps a little extreme with it's distribution of heat. Also 12z GFS for today is 2c too cool for a large chunk of the warmest zone...

Clouds breaking up readily across the south now, expect temperatures to respond upwards. Cambridge region showing multiple pws now at 29c which looks reasonable. Some places will get 30, officially or not from today Broadly the models are ok, though worth noting the 18z GFS yesterday were a good 2-3c too cool compared to current temperatures.

I'd say it makes sense to hold back on the real extreme numbers as it would only need a 50-75 mile shift east and those temps are gone, and that's plenty possible at this stage still. However I'd still count 35 as pretty extreme, even if it does feel like it's getting more common over the past 5 years.