Don't discredit yourself Darren. We all know that there is a little more to this competition than pure good luck. 

Met Office Hadley         18.1c.     Anomaly      2.1c. Provisional to 21st.

Metcheck                      18.18c    Anomaly      1.96c

Netweather                   18.67c    Anomaly      2.48c

Canvey Island              19.9c      Anomaly      1.7c

Darwen                        18.3c     Anomaly       1.82c

Mansfield Woodhouse     18.7c    Anomaly       2.07c 

Peasedown St John         17.3c    Anomaly      -1.02c

Treviskey Redruth         17.4c      Anomaly      0.88c.

Mean of 10 stations      18.26c.     

Thanks Simon - phew just look at those blue bars there.

Such a sorry sight to see all those blues after the summer we've experienced. 

Wondering whether this August CET could still come in under 17C?

As Eddie Cochran sang: "There ain't no cure for the summertime blues."

Spot the bank holiday weekend! Though to be fair, May Bank Holiday was at the other extreme.

For me this UKMO 12z +144 is today's 'hope' chart for those who'd like this summer to rest notably high in the record table of hottest summers. Funny that it comes from UKMO, which was moving the trough furthest east as of today's 00z (and preceding few runs).

The big deal here is that there's enough of a ridge to our E and NE to allow us to entertain the idea of the cool air not being able to make it back across the UK prior to the pressure build from the southwest. That'd make for a much warmer finish to the month, allowing the overall summer mean to sit among the 'greats' of the modern record rather than fall out of that league right at the end.

It's going to take some, though, given the impact this Fri-Sun looks to have. Some unfortunate timing involved there unless things change for the better; if this weekend's incoming warm front was just 6 hours faster, Saturday night would be a lot less cool and Sunday at least a couple of degrees warmer. 

If, instead, we see another load of cool air in and then the ridge builds in and traps that across the UK while winds fall light... the CET would probably fall into the mid-16s, maybe even a tad lower still! 

The August CET is very sensitive to that sort of detail in the final third. The delay of warm weather by about a week has probably made the difference between a mid-high 17s finish and a finish roughly 1*C lower, unless that Scandinavian High turns out even stronger for next Tue and is able to bring very warm conditions from the southeast for the rest of the month.

Using the provisional CET for August, (prior to adjustments of course) and assuming a CET of 18.0 was maintained yesterday (which may not be the case), I calculate a current summer mean of 17.68 from June 1st to Aug 22nd inclusive*. With 9 days to go, the CET for the remaining days  would have 9/83 influence (just under 11%) on reducing the current figure. That means that if the CET is 3 degrees lower than 17.68 in the last 9 days, (say, 14.6) the CET would reduce to about 17.35.

Given that yesterday may have reduced the CET below 18,  the August data is provisional and downward adjustments are likely, (GW has already estimated that it's in the high 17s for August), then it's likely that even that figure for the last 9 days would mean the summer overall drops below 17.3.

In fact, a min of 9 and max of 18 would only give 13.5 as a 9-day CET. It seems most likely at this stage that 2018 will end up very close to, and hopefully slightly higher than 2006.

That's a pessimistic and hypothetical view, full of speculation and lacking plenty of evidence. Current models don't have the nights at or below 9C for the entire 9 days, for example.2003 is still on the cards I would say, but it's still IMO nail-bitingly close.

* [(30 x 16.1) + (31 x 19.1) + (22 x 18.0)] / 83

18.1C provisional to 22nd.

But with this morning's CET minimum down to 11.6, we can see the action of the change in airmass over all but the SE.

No posts from me today. Bee to RUH Bath with the wife having very bad hip pain. Ok now after a dose of morphine. Now in her bed getting a good night's rest.

18.0 provisional to 23rd.

Met Office Hadley       18.0c.    Anomaly     2.1c provisional to 23rd.

Metcheck                    17.99c.  Anomaly     1.76c

Netweather                 18.56c   Anomaly     2.37c

Peasedown St John  17.3c Anomaly  -1.02c.

Mean of 10 stations.    17.9c. 

Several factors now look to be 'worst case scenario' for the CET:

1. The Sunday front not arriving soon enough to keep the preceding overnight temps up

2. The Iberian low moving past (or over) the SE of the UK such that rather than deflecting away the next bout of polar maritime air, it holds it up more across the UK.

3. The development of a stronger ridge over the UK occurring with such timing that warmer conditions take until past the end of the month to arrive.

For what it's worth, the ECM day 10 is what the long-range signals indicated to be unusually high probability for the final week of August, and that was the reason for my high CET estimate for this month:

So you can really see how the delays have undone this month and probably the summer CET's chances of finishing in the top 5.

As far as my own estimate goes, well it looks to end up being unbelievably far out. Back at the end of July, any notion that the month would finish within half a degree of the long-term average was outside of the spread - but that's because the guidance assumed that the natural cycle of tropical activity would play out given the neutral to slightly warm ENSO background. Instead it's been delayed by more than a week, which is most unusual for the setup in place. Signs are, an anomalous SST pattern outside of the tropical Pacific may be the culprit, but this is not certain.

Until this month the long-range signals have produced a lot of good results so far this year, following adaptation of my methods to incorporate the Global Synoptic Dynamical Model (GSDM), but this one mishap looks to hurt the cumulative errors terribly... or at least, those visible in this competition; if the rest of the months turn out well, this one will be analysed as an outlier as part of my own review in early 2019 (therefore omitted from most calculations).

I won't deny, it's annoying that science can still fall so far short of a more instinctive approach sometimes, and I've struggled to avoid falling into a persistent negative mood this past couple of weeks, but I just have to remind myself that what with the nature of chaos, this is always going to be liable to happen every now and then.

You know what, I really miss being able to just shrug my shoulders and say 'never mind, it was just a gut feeling anyway' .

My calculations have the CET at 17.75C up to yesterday.

Latest estimate of the final August CET is 16.67C.

That would leave the summer CET at 17.31C which is 5th in the series. Currently we are still in second place but will drop below 1826 tomorrow.

My money's on 5th place, but I still think we could end up ahead of 2003. To fail to beat 2006, I think the CET would need an average below 14C for the remainder of the month. A run of unseasonably cool days like that does happen- 2014 for example; but I still think the models suggest a few interruptions to the very low CETs and enough to get towards 17.3 for the summer. Hope I'm right- I'll not be disappointed with 5th.

Cloudy breezy nights will not get too low.  Its the calm clear ones to avoid, as we know.  The forecast is very variable at the moment, perched as we are between high and low pressure.  All to play for. August CET could still end up 17c+,  IMHO, with promotion in the Summer CET series as a bonus. 

Thanks GW! 

That's quite a drop for week four isn't it?  

Still Summer 2018 comes out as a "top five" Summer so we can't complain with that...

We shouldn't complain but we should feel bitterly disappointed by the crash in the last week.

I guess this is the start of the decline towards the long hard frigid winter of 2018/19. 

Cheers GW, looks like I was being a little too pessimistic with my CET estimations and we could still finish in the top 5 for the summer CET, which would offer some consolation for the collapse of the August CET.

It really is extraordinary to think about how much of a drop is on the cards considering it's the final week of the month. I wonder how exactly that stands historically in terms of final week drops... at least among Augusts? I imagine there are winter months which have fallen further via potent cold spells.

1st day of the new month  

The weather can't really troll us... can it? 

For all my plain scientific logic, the way things go over the coming week actually creeps me out a bit.

It makes you wonder if September will be like 2006.

Such a shame summer has totally collapsed in August again!!! To think that if we had had another blast of heat now instead of these autumnal temperatures we might have topped 1976.So near yet so far awaythe August curse strikes again!Oh well I guess 3 scorching bank holiday weekends in a year is too much to expect. Summer proper was definatly May,June,July and first week of August this year an epic 3 and a bit months which will live long in the memory but at the same time a bit disappointed at this huge crash.

My estimate for summer CET stands at 17.37 currently.  August to come in around 17.0, all subject to change. Fascinating.