Ops run continues its bipolar behaviour

i find it extraordinary these highly sophisticated models are not able to agree on the potential heat just 5-7 days out. For many people especially the elderly this is a big issue. One extreme to the other....

To be honest I still find it more extraordinary how accurately they can forecast a chaotic atmosphere several days ahead!

UKMO 12Z looks fairly hot: it has a 36C isotherm on Monday 22nd, and bear in mind it forecast a max of 25C today, when Heathrow actually exceeded 27C.

It's not 'chaotic'*, it's behaving according to the laws of physics.

*not in a state of disorder. It is clearly 'ordered' in that it's not going to be -1C tomorrow and snowing...I know a river will flow down hill, the current direction in that river, the eddies, more difficult to predicts. So the deep signal for heat has been there for some time in the models, the eddies less certain.

At least that's how I see it. So I think the models are showing the heat dome slowly relenting as we go to July. My hope is the westerlies return because I do genuinely fear a 'roast, simmer, roast' scenario...

The 12z GFS Ens were not hugely different to 6Z in terms of overall average 850Hpa and rainfall amounts in the south east. Apologies for the IMBY post ..... 

Yes GFS  continues its struggles.

AIFS ensembles are more solid though. Waves of heat hitting the South. GEM is similar. 

https://wetterzentrale.de/en/show_diagrams.php?geoid=49069&model=aifs&var=2&run=0&lid=ENS&bw=1 

Yeah, hot outliers.

A cooling trend in the EMS of sorts.

Still a massive spread though. It shows how sensitive the conditions are to a slight shift in the position of the high. If it nudges west just a small amount then it allows a more northerly drift to push the heat back into France. 

Still, even the cooler options look fairly HP dominated at least for the south. 

Talk about convective.🌩️

According to TORRO the all-time UK record is 578dam, and that was actually in June - June 28 2019. https://www.torro.org.uk/extremes/summary

The 0Z GFS has 579dam around the Herstmonceux region here: https://www.torro.org.uk/extremes/summary

Indeed; 100mm + ppn over parts of central England.

As early as next Tuesday GFS has a 20 degree spread of maxima for London from 18C to 38C. That's only day 7.  Days 4 and 5 have a spread of 7 and 9 degrees...

That's not an honest reflection of the range though is it? If you take out the 3 or so obvious anomalies in the upper temperatures for example they barely have a range of 10, not unusual. The surface maxima will naturally deviate after day 5 and even they barely manage a 15C variation at day 7 once extremes that are not part of a group or have daylight between them and the nearest group are removed.

The trouble at the moment is a tiny shift on a geographical scale of the boundary between the continental heat dome and our more benign warm HP can result in 100s of km difference. Each time a new factor is injected into the models, differences of that scale become likely.

This was the case, the 12 z GFS now has the perfect scenario for UK wide heat middle of next week with high pressure strengthening to the north and east of the UK.

The 12z GFS continues the trend from the 0z in terms of moving the ridging north - which means my area escapes the worst (due to an onshore breeze) and the 20C 850 isotherm is safely kept at bay. 

Don't look at the UKV then - 22C at 850mb over Sheppey on Monday. Although you would get a sea breeze.

TBH the way the models are struggling with the boundary of the frankly tropical airmass versus something very warm instead reminds me very much of winter, when they struggle just as much with say the -10 850 line. In recent years the struggle has always resolved, in winter, in favour of less cold conditions - the same doesn't apply with summer heat.

It does look like there will be an onshore breeze down here from Sunday onwards, and that will keep the far SE that bit cooler. Further inland, though... 

The MetO text forecast mentions isolated thunderstorms, but the UKV output would suggest more general showers - it's lacking the intense rates that you normally see when storms are forecast. The MetO raw has a 60% chance of light rain in the far SE during the early afternoon, it'll be interesting to see which model ends up being correct!

EDIT: The MetO raw now has a 90% chance of rain (at 25C too, which is uncommon), and a good deal more cloud around too (not that I'm complaining). Clearly something's been picked up at the last minute...