Mark Smithy

This could be a month of two starkly contrasting halves. It seems extraordinary right now to contemplate it as we a) have a general tendency to recency bias and b) meteorologically high pressure patterns can be hard to shift. But there's no doubt that the models are now pointing to a significant change:

GFS overbaked today's cake by 2C pretty much across the board and with its operational still out of line with almost all of its own members and all the Euro models, it looks like this time it is overstating this week by around 2C for the majority. Some isolated places may reach 35C or 36 but they will be exception. Looking ahead and there is now good model consensus amongst the Euros and most GFS members for the breakdown to start Sunday evening. As I have been stating, the devil will be in the detail for precipitation but the omens are excellent for some much needed rain. The ECM is a thing of beauty for those of us who would prefer now to see parched ground revived, rivers filled and wildlife given a chance once more:

But if anyone disagrees that these models will produce significant rain over the second half of August I'll happily bet a tenner, proceeds to a wildlife charity trying to help our poor animals suffering through this. I don't think I've used the word cool? Unstable airmass and low pressure evidently now showing on the models. It will be significantly cooler than the next 4 days though, even if the GFS this time is overbaking the cake.

An example of the devil being in the detail. A recent ECMWF chart showing potent precipitation in the south and a latest FAX chart showing a trough. With this degree of instability rain will be inevitable. But when, where and for how long will be down to hour by hour on the day. Ever thus.

This is wide of the mark. It's akin to the 'will it snow in xyz' type posts in winter. The key thing is the setup. The fact is that we are moving to a significant low pressure phase next week and unless you are pinning all your hopes on the lone star GFS operationals the devil will, as always, be in the detail. With the low pressure will come precipitation. Significant quantities of it. The Euro models all show this and so do many of the GFS ensemble members. We have a great tendency to recency bias. We subconsciously think 'oooooh it's hot now it will be hot next week.' Actually the models are consistently showing us that the second half of August will be a marked contrast to the first half. As the Met Office have said.

If this was mid-winter I'd be stating that I don't trust the GFS operational one iota. It is continuing to be out of line not only with the Euro models but also with the mean of its ensemble members, significantly so. Here it is again on the 18z and the 0z. That's almost a week now where it has been overbaking the heat. For cold lovers this would be a red flag. It might of course be right, but it has very little support from both within and without.

Interesting situation with the GFS at the moment. As we know, the Euro models incl Fax charts have a relatively less extreme hot week and a faster breakdown into something less warm and much wetter next week. So do the majority of the GFS's own ensemble members. But not the operationals. For the past 3 runs, and pretty much for the past week, the operational has run far hotter than the mean. Here are the last 3 18z, 0z, 6z: My advice is to be wary of pinning too much on the GFS operational runs at the moment until or unless it comes into line with everything else. There is very little support for 1. its extreme version of extreme heat this week 2. slower breakdown and 3. return to heat and high pressure within a day or two. Not unless you pin all your hopes on an out-of-kilter operational. See my post.

0z ECM precipitation chart at 7 days: And 10 days: Euro models seem to be in agreement at the moment that after this week's extreme heat a return to something more typically British may be in force from next week onwards. This tallies with the Met Office forecast for a wetter second half of the month.

Two charts from UKMO and GEM that are not stable summer weather. In these situations I would always trust the European models too. We await this morning's ECM with interest. Especially so as the GFS has habitually over the past week had a problem with its operational being out of kilter with its ensemble members. A swift breakdown coming end of the weekend.

I think the br You misunderstood. I meant that the breakdown when it happens will be swift. As I also stated, we have a week of extreme heat to endure first. The two statements are complimentary not antithetical. In terms of precipitation, the set up will certainly produce rain in the south. The devil is always in the detail and impossible to forecast at this range. The main thing is that we are transitioning from dry heat this week to low pressure next week. Which for many of us (but not all) will be very welcome.

The models are now aligning on a fairly swift breakdown to the latest heatwave, with next weekend seeing low pressure troughing. 0z UKMO, GEM and GFS all now agree on this and, one assumes, so will ECM out in a little while. Just some extreme heat to endure first this week, then we should be back to somewhat more normal British summer conditions. Although the devil will be in the detail, all of these scenarios should eventually lead to the much-needed precipitation in the south. The Shah of Iran used to call this country's rain "liquid gold".

Despite the 6z being a less hot run than the 0z it is still, surprisingly, a bit out on a limb. This fills me with great hope that the GFS operationals have all been over-egging the heat. The ensembles seem to be onto something with the breakdown, which is a clear signal at day 8. When this happens in winter it sometimes pressages a faster breakdown so I am also hopeful that the damaging extreme heat will be swept away sooner than a week hence. Considerable precipitation revolving around the low pressure in the second week if this run came off but that's very much FI.

The 0z GFS is very much on the high side for next week's temps amongst its ensembles. This gives me a glimmer of hope that it may not be as bad as the heatwavers think. Probably more in line with the Met Office. And thankfully, nothing like as bad as last month (at the moment): Further ahead, the cool northerly is an outlier as is the FI return of heatwave. This pattern of the operational being out of kilter with the mean and control has been going on for some days. Might be best to look at the GFS ensemble mean.

ECM 10 days away. Fingers massively crossed that this comes off. Sorry to the UK staycation holidaymakers but we desperately need rain and there's little cause for celebration about the extreme heat.

The outputs seem very confused at the moment. The high heat has returned for mid-next week for c. 4 or 5 days (argh) but although all of the models seem convinced the HP will move, none of them seem quite sure of how. The 0z GFS returns to its bifurcation theme with a potent northerly (polar low): The 0z UKMO has the high shunting off to Poland which I suppose might lead to a northerly trough: One thing few of the models have at the moment is significant precipitation. Given the strength of winds that some of these set ups suggest then serious wild fires may ravage parched southern Britain. Imho this is no longer something to celebrate but to be concerned about. Grim times: for reservoirs. For rivers. For nature. For agriculture. For lawns. For the planet ... which is burning up.

A glorious northerly in mid August. I'm personally loving the downgrades. We're not built for extreme heat in the UK. Our homes aren't designed for it and if I wanted that, I'd live in another country. Really looking forward to autumn, a season I love, and now there's just a hint of a bit of freshness in the air from time to time in these runs. I love that in August: how it switches from a cool morning when you think 'ooooh, that's a hint of what's coming' but then switches back to beautiful warm days way into September and even sometimes in October. So nothing crazy now showing on the models for next week. Bit of heat. Bit of freshness with the HP splitting midweek and troughing. It will also be very interesting to see where the GFS operational sits in the ensembles. There are BIG splits now occurring with a number of much cooler options showing:

It has been beautiful today. Gorgeous sunshine, lovely breeze. Nothing crazy. Just a lovely British summer's day. And as for "7-8 months of cold" that will be news to all of us snow and cold lovers. We barely get a day like that nowadays. Autumn is often beautiful and mild and rain is what makes this land green and pleasant rather than an arid desert.

The 0z ECM also abandons the boiling hot ship with a bifurcation in the high midweek and a much cooler trough pushing down across the UK! Phew! Could be lovely with gentle, benign, traditional British summer weather. I'd also be thrilled with continuous heavy rain for 24 hours to give our gardens much needed drink and our reservoirs and rivers a chance to top up. In southern Britain. The UK is a fairly large place, you know

Interesting synoptic battles, or confusion if you prefer, in the model outputs at the moment. The GFS 18z was a hot run but also has significantly less support than previously, which hinted therefore at what was coming with the 0z GFS: And, sure enough, the 0z GFS is considerably less hot than the 18z. See, for example, this comparison: 18z: 0z: Essentially the models are undecided about the trajectory of the high: an eastward movement or a split? The T120 FAX chart for Monday seems to hint at the latter option, although the 0z UKMO doesn't until T144 when it dangles the prospect: one of those charts where you wish it could roll on just another 6 hours. As someone who a. desperately wants rain in the south and b. doesn't enjoy high temperatures unless there's either aircon or the sea, I'll be very, very, happy to see the slightly cooler outcome as per the 0z GFS. Sorry to you sunny lovers. Actually I wouldn't mind hot days and cool nights. It's the plumes I can't stand without proper infrastructure. And we need rain.

In southern Britain

Well there's definitely a high pressure signal in the runs. But the ECM has it heading north, a perfectly plausible scenario with the jet buckling. The last of those ECM charts is T240, so hardly reliable, but would be a setup to bring much needed and glorious rain to the parched southeast. Those low pressures moving up from France are often excellent for intense precipitation.

Yep, the high has been a bit flaky on some of the recent model runs. The 0z GFS returns it though. The 18z was on the lower end of the pressure ensembles: However, the 0z UKMO also looks a bit sketchy: The underlying 'problem' is some mobility in the Atlantic which is pushing low pressure cells across to our north, and high pressure cells moving west to east, as has happened this week.

There does seem to be a decent high pressure signal about 10 days from now which, were this winter, would not be considered reliable. There's a relatively strong jetstream for the time of year which will have to buckle north for the Azores high to build: And before it does so, I still maintain that there's a fairly cool north-westerly to kick through first: Until at around T220 we arrive at this settled high pressure scenario: In the interests of balance it should perhaps be mentioned that the GFS ensembles show the operational to be marginally on the warmer side at that relevant timeframe: But the signal for high pressure is reasonable, albeit not exactly a very strong HP cell showing at this stage at around 1020 mb on the operational: I'm not suggesting the return of high pressure is unlikely but neither is it nailed on.

Erm, what am I missing here? There's still some cool-ish conditions showing on the 12z GFS later next week?

Cold night for the time of year with some temps dipping as low as 2C or 3C. That'll knock a decimal point off the CET.