GFS Op is quite extreme for the Central Highlands. Although the ensemble mean runs 3-4 degrees up on the numbers listed here, there is support for the early cold from several Ps. 

12 noon temps from 22nd Sep to 4th Oct inclusive go:

10   4   5   3   3   3   5   7   5   6   10  5   5

Several snow occurrences, with a biggie suggested for next Thursday, may be thin probabilities, but some low temperatures UK wide seem to be likely at least at the start of this period.

Yes, it certainly looks like a bit of a cool down in the final week of the month. Nothing too unusual but a flip from warmer than average to colder than average will probably feel quite pronounced. 

Good stuff with the resolution increase Brian. It should come in handy in the winter.

Raging jetstream to the N progressing slowly south to Tue 22nd when knocked into a loop by influence of Teddy and enclosing the UK in LP Thu 24th. The loop moves E-wards only slowly so England is under a strong flow from NW on the rear side of the loop which continues though weakening to Sat 3rd. Signs of a new loop then developing so still unsettled.

GFS Current generally HP with LP to south disappears and deep LP approaches Thu 24th  moving E-wards slowly and leaving a legacy of N-ly gales. Another LP moving in from N Atlantic  to N Sea 990 mb Thu 1st . Brief rise of pressure Sat 3rd but more LP on Atlantic waiting to move in

GEFS  Like yesterday, temp drops sharply from mid week ca 7C below norm Sun 27th, recovering uncertainly to norm at 3rd (op & control agree  but plenty of variation in other runs) . Rain Thu/Fri 24th/25th, not entirely dry after that but no clear pattern  in S (Scotland more likely around Sat 3rd) Snow row figures for Inverness up to 7, even Glasgow gets a 1 on 28th Sep.

ECM (largely 12z) Not such a deep LP nor N-lies from the Arctic on 24th/25th, and quite settled by Mon 28th though with LP lurking near Iceland for later that week EDIT Much the same on today's 0z

Close to an ice day in the Scottish Highlands on Sunday 27th if GFS 00z is right. (Probably won't be this cold!) https://t.co/4nZiyFawPd) pic.twitter.com/ysPTlC2ws9

— TheWeatherOutlook (@TWOweather) September 19, 2020

Weather does look very poor from about the 23rd for at least a week. Not good timing with Covid19 kicking off again as well.

[quote=Ally Pally Snowman;1256866]

Weather does look very poor from about the 23rd for at least a week. Not good timing with Covid19 kicking off again as well

Long social distancing queues outside for a bag of sugar ....wont be so much fun with no warm sunshine

 Well it depends what you expect really. September has been predominantly dry and warm and even with a marked cool down will probably end up another warmer than average month. I am quite looking forward to a bit of cooler air in the mornings as it has been t shirt weather for ages.

 As for the Covid -19 it has started to get worse earlier than i thought it might, so not right in the depths of Autumn /Winter.

Well it had to happen I guess, the change in the weather I mean. Law of averages and all that. I suppose you could say the same thing about the virus as well - that was always unavoidable too.

Can't complain really.

We have had a great summer - esp here in the SW been dry and warm since lockdown on 23rd March.

Meanwhile looks like the -5c isotherm is almost clearing the country by the end of next week into the weekend! It's rare to see -5c @850hpa in a September month...would you not agree!?

Using weatheronline as a guide, for what it is worth.

If the temperatures pan out as predicted from Sunday 27th (a long way off I know)

The last time when temperatures of 13-14c have occurred at Heathrow in September for more than 4 days in a row was 1994.  Even then it bounced back up to the 20's.

I noticed the 18z GFS are showing single digit max temps for my neck of the woods by next Sunday. It will get modified as we get nearer to the time but that would come as a shock if that actually bore fruit.

GFS - Last of the HP tomorrow, Mon 21st, declining first to W-lies then LP 995mb Ireland Thu 24th crossing the UK eventually finishing up in Germany Mon 28th with ridge of HP moving in (but not before delivering some cold N-lies). Then more W-ly/NW-ly until Sat 3rd when LP 980mb N Sea delivers a fresh batch of N-lies.

GEFS - Dip to cold Fri 25th with plenty of rain around that date. Mean of runs slowly recovers to seasonal norm Thu 1st but a great deal of uncertainty (the op in particular doesn't know what it's doing) and some rain on and off continuing to end of run Tue 6th Oct. Temps in Scotland more stepwise - stay colder for a few days longer then recover more abruptly to a little above norm. Less rain there but by no means absent. Snow row figures for low altitude locations now back to zero.

ECM (yesterdays's 12z) - More of a W-ly feel than GFS; next weekend's LP less deep but equally ridge following much flattened. New LP appears Rockall 985mb Tue 29th, not shown on GFS - UPDATE- next weekend's LP now deeper to match GFS though no ridge to follow and the LP later on is 985mb N Ireland Wed 30th rather than staying out by Rockall

I'll eat my hat if that actually happens!

0z run looks chilly and unsettled 

6z gives us a cool unsettled start to October

9 degrees  Celsiusdaytime max  showing for London Sept 30th.....remembered it before like that nearly 30 years ago....that was followed by a mild winter...

Since that's ten days away, should be taken with a large pinch of Na Cl.

Divergence in the models tonight. The GFS hints at the chance for quite a widespread frost on Saturday and Sunday morning especially, however the ECM brings milder air in on Saturday, which would of course stop this from happening.

Could be quite an interesting period for forecasters, with so many ex-tropical systems knocking about in the Atlantic over the next week or so.

Jetstream buckling to produce loop around the UK Thu 24th; as this moves away the stream straightens out across the UK but furter loops develop in vicinity of UK Thu 1st and Tue 6th

GFS - LP over UK Fri 25th not as deep as previously forecast, followed by W-lies for a while before LP over Scotland Thu 1st 985mb decaying and moving S to Biscay 1000mb Tue 6th

GEFS - temps dropping to well below norm thu 24th, recovering to norm Thu1st but with little agreement between runs by then and thereafter. Less rain in S this week than previously forecast but with quite frequent big  spikes in different runs from Tue 29th onwards, less dramatic but more continuous in Scotland

ECM (as of yesterday 12z) - similar to GFS but LP Wed 30th (end of run) not as deep as GFS ALL DIFFERENT THIS MORNING'S 0Z - Lp on Wed 30th banished to Iceland

16-dayer cool across all NW Europe for a week, recovering slightly here but focus of cold moveds to C Europe later. Swathe of rain Nw Britain to Romania.

GFS - W-ly becoming NW-ly/N'ly from LP moving across UK with ridge of HP just about making into the W Sun 27th but collapsing to more W-lies; trough of LP developing over Iceland and extending into N Sea Sat 3rd; back to more W-ly before HP arises covering UK Thu 8th. 

GEFS - cold (7C below norm) Sat 26th but back to norm Tue 29th and staying there with quite a lot of variation between runs though less than yesterday (one monstrously warm outlier 15C above norm 6th Oct!). Rain around in next couple of days and again 1st - 3rd and in some cases 7th Oct. Scotland same tendency but missing that outlier and with less rainfall, some runs pretty well dry throughout in E.

ECM - Similar to GFS but W-lies look warmer around 1st Oct

GEFS gets a massive production upgrade tomorrow. Improvements include:

1) Higher resolution

2) Use of the FV3 core (used by the latest GFS)

3) 30 perturbations + control run

4) 00z cycle extended to 35 days

Today on TWO the 12z GEFS update is running the new version which is available on the parallel NCEP servers. However, there will be 20 perturbations + control run as normal. 

Later this week I will hopefully make available:

1) 35 day plots

2) All 30 perturbations  + control run

I would expect the GEFS updates will be slightly delayed due to there being more and bigger datasets.

12z GFS keeps us cool and unsettled up to the end of the first week of October 

The rain will do the gardens some good 

Jetstream in looping mode for this week, the loop transferring across the UK with the trailing edge giving a N-ly blast on Friday before modulating into a tentative sort of W-ly Mon 28th with  the main thrust running from Newfoundland to Spain by Fri 2nd i.e.just W of UK,but with occasional fragments swinging back N-ward over the N Sea. These become larger and eventually form a cut-off circulation enclosing the UK Mon 5th. This breaks up chaotically, but there is a branch running NE-wards over the SE Thu 8th

GFS - LP moving SE-wards, in southern N Sea by Fri 25th; then a W-ly pattern for a few days before LP covers England  980mb Sat 3rd; this hangs around and slowly fills, remnants 1005mb over Cornwall Fri 9th. No strong correlation with jetstream pattern after the next few days

GEFS - agreement on temps v low Sat 26th (8c below norm); back to norm Tue 29th and the mean staying there through to Fri9th but with quite a lot of variation after Sat 3rd. More rain than forecast yeaterday and frequently from Tue 29th

ECM - the new LP arrives earlier (Thu 1st) and drifts off to Norway Sat 3rd rather than filling in situ.