Yup. GFS like a dog with a bone on possible heatwave conditions again later next week.

ECM/UKMO remain in the cooler camp, so at the moment, it's anyone's guess what we will be seeing later next week.

You mean like the current depressed cool and unsettled conditions?

GFS sticking with the repeated pattern of high pressure building from the south with warm upper air (high teens at 850mb) then temporarily being flattened out by Atlantic lows that never really make make much of a dent in the high and get deflected off to the northeast. The result being heat building from mid 20s to low 30s, then a brief dip to lower 20s, rinse and repeat. It can keep doing that all summer as far as I am concerned 🙂 

Sorting the GEFS members by temperature gives a nice visualisation on the table view on Meteociel:

https://www.meteociel.fr/modeles/gefs_table.php?x=0&y=0&lat=51.5085&lon=-0.12574&run=6&ext=1&mode=7&sort=1 

definitely another peak centred around the 28th-29th with the median 2m max in London being about 26C.

P13 is the obligatory scorchio run, hitting 39C.

Seems pretty accurate even if the heat may have been overcooked. Highest temps currently are up in the northeast with Albemarle and Newcastle airport on a rounded 30C! Still lots of cloud around but it's now Ac castellanus rather than the featureless cirrostratus.

I saw it come out and was waiting for the post. 🙂

Well I'm not wrong am I? Yes, one run, one model, but looks really wet later next week, and cool or even very cool.

You weren't wrong, but now you are. It is not really wet and neither is it very cool.

The precipitation chart on Meteociel suggests otherwise. 20-30mm over a 24-36 hour period is pretty wet in my book.

Yes in NW Scotland, The lake district get 20-30mm, but mostly from TStorms tomorrow. The rest of the country 20-30mm accum over the entire run.

Still no trend from GFS which is also, like ICON, much cooler now for next week.

The cross model trend is definitely cooler and more changeable, but not unusable weather. 

However, once this weekend is out of the way, we are I think looking at a spell of temps at or slightly below average, nationwide.

For my location the max temps from tomorrow until next Sunday run 30, 24, 22, 25, 24, 22, 24, 29, 32.

10-day rain chart isn't out yet but the 5-day one has 1mm accumulated ppn out to Wednesday.

Bitter in the wind probably some snow on hills. 😁

https://www.wetterzentrale.de/en/topkarten.php?map=20&model=gfs&var=17&run=12&time=219&lid=OP&h=0&mv=0&tr=3#mapref 

UKMO firmly in the ICON camp, unsettled and much cooler next week.

GFS looking increasingly isolated. 

Furnace heat again on this afternoon's GFS. 

Very isolated though Brian when weighing up the other models.

No other model has got anywhere near this. GFS is not having its finest hour just lately.

Hang on, the other day you pointed out the folly of criticising something for what hasn’t happened yet in the future??

GFS has performed rather well over this spell actually. Change the record 

Tbf UKMO is rubbish  but I never rate it's 144h/168h charts.

To apply a bit of balance, GEM 12Z also keeps high pressure and warmth in charge right out to the end of the month.

I suspect as often the case the reality will be somewhere between the two: not the 15C and unsettled of some models, but also not the 34C hairdryer of the GFS.

A halfway house of low to high 20s in the south and plenty of sunshine with perhaps the odd showery day, then.

Decent AIFS run again, brings in the heat at 168h. It's been very consistent with this. It is also the best performing model atm

I see that the BBC TV forecast had the temperatures back up to the high 20s in the SE by the time we got to the end of next week. This probably reflects the model output quite well. It will be cooler and probably more unsettled at times elsewhere, especially in NWern parts.

I agree, I'm discounting the GFS runs in the extended range at the moment based on recent performance it seems to be way overdoing the heat.

It has been a bit over the top with the actual temperature but the synoptic pattern has been pretty accurate.