Indeed. The BBC are still in their pictorial suggestion, predicting 4 hours of light rain tomorrow. On the other side, the Met Office have given up completely on any rain in their pictorials for here. Both forecasters in their written predictions suggest isolated thunderstorms tonight but based on past events I won't expect much. Hope to be proved wrong of course. 

GFS Op 00z; a balance between three sources - HP from the SW (fine and sunny), LP from the Continent (exporting hot and thundery weather) and LP on the North Atlantic (cool and breezy W-lies). Tonight a touch of the hot and thundery in the SE (latest forecast is Kent and E Sussex only but it keeps changing), then a couple of days of Atlantic weather. From Sat 28th for a week mostly the HP (with the Atlantic affecting the NW), then from Mon 7th, the heat dome on the Continent spills over into Britain before the Atlantic (with aid of a heat LP from Spain) regains control. But only a slight shift could give any one of the three the upper hand.

ECM; similar to GFS but the HP in the week from Sat 28th is weaker; more Atlantic influence esp from the NW at first and more Continental influence for the SE later.

GEFS; In the S, mostly hot to Tue 1st, the a quick drop to normal temps and the mean staying there though a few ens members led by the op go for heat Tue/Wed 8th/9th. Very little rain and that intermittently. In the N, mostly warm to Tue 1st but not consistent; then a drop with temps below norm to Sat 5th, followed by a slight increase (the insanely hot op on the 9th is here, too, but out on its own). Rain likely around Tue 1st, small amounts at other times.

AIFS; London - steady rise of maxima to 32C Mon 30th, then maxima in low 20s through to Thu 10th (note that Tue 8th maximum actually below 20C, a stark contrast with GFS and a good test case), some rain Wed 2nd and a bit more Tue 8th. Edinburgh - temps also rising to a maximum Mon 30th but more irregularly and only reaching 25C, then the drop with most days' maxima close to 15C. Dryish Sun 29th to Thu 3rd, small amounts of rain at other times.

6 runs on the GEFS bring 20C+ upper air to London (including the op run, in a big way!)

Certainly looks like there has been an uptick in support for another hot spell around the 7th to 9th of July on the ensembles, but a long way to go.

Before that, there is still considerable uncertainty in how quickly the hot air at the end of this week will be swept away. Some models clesr it through on the 30th, others keep it in place until the 2nd. (One rogue GEFS pert even has a second spike on the 3rd!)

Quite strange conditions this week - it's looked cool and grey from inside, but then when you go out you realise it's actually still really warm!

And the heat from saturday still trapped in the house. having to sleep with a fan still, feels muggy and horrible. 

obviously a different story further north and our weather divide between north/south - especially south east/east angular being almost part of mainland europe in terms of our climate

This weekend and beyond look awful again for coldies like me 

Another GFS op run casually throws out 100F, that's two on the bounce. "Only" 33C on Monday though before the cooler air pushes in.

Worth noting though that the latest GFS looks a bit wrong even at just 12 hours out - it's suggesting 28C across Kent by 1pm.

Although actually looking at current readings there is a big gradient across the south - just 21C here but 25C at Charlwood.

As early as Monday, the spread of maxima for London is from 21C to 37C.

By Wednesday, it's from just 15C to 33C!

ECM ENS suggests 30C to 34C on Monday as the max in the London area.

21 C in Scotland's Capital 🤔

I'll swap houses with you Beast! House temperature between 18-20 C here.

Outside it's around 18-20 C too 😁

AIFS has 32c Sunday, 35c ish Monday when adding on the usual 2c then probably 29c Tuesday.

  It's then more unsettled than GFS.  Don't bet against AIFS at the moment it's on a roll.

GEM peaks the heat at 35C on Tuesday 1st.

UKM global delivers a sneaky 35C on Tuesday. That's about par for the course in this day and age. 😂

They have had a hot start to Summer in parts of the U.S. and the warm airmass is here now, well that is what the BBC say in their shock, horror piece on the coming heatwave , which atm only looks really hot in the extreme east.

After the heat AIFS again not interested in any significant HP building in , in fact it goes very unsettled. Will it be right again?

I will be back by 10th July and this chart extreme heat will greet me if persst on the10th. I think this summer going to be warm to hot side.  Even today unsettled weather with some sunshine and strong winds still deliver above average temps of 24-25c compare to 14-15C last year unsettled non-stop spells.

The AIFS has been pretty rubbish here at least, for want of a better word.

Here's a reminder of what it was showing for today a week ago:

https://www.meteociel.fr/modeles/ecmwf_aifs.php?&ech=168&mode=19&heure=12&jour=18&mois=6&annee=2025&archive=1&carte=6&aifs=2 

A high of 22 today (in reality 27), a high of 21 tomorrow (in reality 25 is likely) and so on - 23 over the weekend, when it'll be close to 30. 

Even 4 days ago it was woefully wide of the mark:

https://www.meteociel.fr/modeles/ecmwf_aifs.php?jour=21&mois=6&annee=2025&heure=12&archive=1&mode=19&ech=168&runpara=0&carte=6 

Just 21 today, 22 tomorrow and 24 for Saturday.

It's just glorified pattern matching, and while it has higher skill at the 500hPa level than the traditional version, it's still just as prone as they are to fall for the "fake cooldown" scenario, i.e. over-egging the jet. 

Temps I agree it's always 2 or 3c below for some reason. But pressure patterns it's been the best recently.  It picked up this weekends heat 2 weeks ago.

Be very interesting if it sticks to the unsettled pattern and if other models follow 

It's more than 2 or 3C below, and just as the other models have done it's been crap with its handling of the jet.

As for 2 weeks ago?

https://www.meteociel.fr/modeles/ecmwf_aifs.php?jour=11&mois=6&annee=2025&heure=12&archive=1&mode=19&ech=168&runpara=0&carte=6 

It showed 25 for Saturday here and just 19 for Sunday, I'd say that was a pretty bad forecast at that range... as you'd expect, really!

(GFS has 31 here on Saturday and 32 on Sunday, FWIW, MetO has 29 on both days).

EDIT: Not to say the AIFS is completely useless, it isn't, and considering that the entire output is generated in just a minute or two versus hours of crunching from the normal ECM, it's hopefully something which can be tweaked and improved, but it's not a panacea. It failed just as often as the main models did last winter, too, and frequently flip-flopped just as they did. At least some things never change! 😂

Most good summers have commenced with a  crap June apart from 1976. I think the mechanism is in place for a record summer. 🙂🙂🙂🙂

Pressure patterns look decent enough from 2 weeks out. For me it's been head and shoulders above the other models this summer. Doesn't mean it's always right though. 

Do we have any statues? usually each model is better at different things ie gfs has a natural refusal to undercut but good at picking out a potential pattern changes also good in the past at picking out a northerly.

ecm on the other hand tends to overdo pressure at Greenland as does the Japs

ukmo often tends to be sensible more often I note the middle ground but has difficulty in picking out a trend.

as many have said over the years often the blended solution Is the right one.

Someone posted some stats on  Netweather a few days back and for 500hpa at 5 days it was in front.

Just ahead of ECm with those 2 way ahead

I thought I'd posted them last week, but here you go:

https://charts.ecmwf.int/products/plwww_3m_fc_aifs_wp_mean?area=Europe¶meter=Geopotential%20500hPa&score=Root%20mean%20square%20error 

IFS is just ahead of AIFS at the moment on the European 500hPa grid.

...and AIFS is really struggling with 850s compared to the main model:

https://charts.ecmwf.int/products/plwww_3m_fc_aifs_wp_mean?area=Europe¶meter=Temperature%20850hPa&score=Root%20mean%20square%20error 

I suspect that's due to its database being full of charts from the old days, when the 20C isotherm was pretty much unknown, when -10s in winter weren't remarkable, etc.

Another roasting GFS this morning breaks the June record on Monday with 36c. Then ramps up the heat again soon after. Both GEM and UKMO also look like they would break the record Monday.