This thread is for all comments, discussion, and analysis of temperatures in the UK during September, with particular emphasis on the Central England Temperature (CET), for the simple reason that it is the longest-running temperature series in the world with over 360 years of data. But you can comment on any interesting temperature statistics or data from across the UK, including your own backyard.  

For those taking part in the competition, all CET predictions for September should be sent directly to me via the forum private message system. Please do not post them directly into the thread.    

Historic CET summary for September

1981-2010 (30 years) 14.03C

1991-2020 (30 years) 14.17C 

2001-2020 (last 20 years) 14.27C 

Last year September was average at 14.0C. Actually the Setember CET  has been very similar and very close to average for the past 4 years varying between 13.5C and 14.3C. 2016 was very warm with 16.0C and 2015 was cold with 12.7C.

Here is a chart of the September CET for all years since 1961   

Direct link to a larger version of the chart

Current model output     

The first week shows 850s above average but that is deceptive due to lots of cloud. So unlikely to see surface temperatures much above average - mean of 21C for a few days in London. After that the 850s are close to average with a lot of scatter. 

GEFS 

https://www.wetterzentrale.de/de/show_diagrams.php?geoid=49069&model=gfs&var=2&run=99&lid=ENS&bw=

https://www.wetterzentrale.de/de/show_diagrams.php?geoid=49069&model=gfs&var=5&run=99&lid=ENS&bw=

ECM (de Bilt) looks similar with temperatures struggling to push above 20C at any point.

http://www.knmi.nl/exp/pluim/Data/PLUIM_06260_NWT.png

Met Office contingency planners outlook  

https://www.metoffice.gov.uk/binaries/content/assets/metofficegovuk/pdf/business/public-sector/civil-contingency/3moutlook-son-v1-secure.pdf

A 50/50 split between an average and a warm September based on the probabilities. The Met Office had almost the same split for August and we ended up with the average figure. I rather suspect the same will happen in September.

Precipitation also looks close to average. Another rather nondescript month in the offering I fancy.

A quick look at an interesting statistic. The CET figures for the last four months are remarkably similar to 2010. 

Year            May      June     July      August

2010           10.7C   15.2C   17.1C   15.3C

2021           10.1C   15.5C   17.7C   15.8C

The only other year with a similar pattern is 1994 (although June was 1C colder than 2021). Nov 1994 was of course a record CET and it was followed by a very mild winter. My gut feeling is that this year will more closely follow 2010 than 1994, although I do not expect a negative CET in December.

September 2010 saw a CET of 13.8C. I expect 2021 to come in somewhere close to 14C.

Just seen it - stickied it.

Be interesting to see the spread of guesses/methodically arrived at predictions this month. Things are looking a touch warmer for the first 7 days this evening than when I made my guess, makes me wonder if I shouldn't have gone a bit more ballsy and seen if I could fluke another unlikely result

Met Office Hadley     14.4c      Anomaly    -0.5 provisional to 1st

Metcheck                  14.22c    Anomaly    0.50c

Netweather               15.44c    Anomaly    1.45c

Peasedown St John   14.31c    Anomaly   0.03c.

Met Office Hadley      15.1c      Anomaly      0.3c provisional to 2nd

Metcheck                   14.65c    Anomaly      1.77c

Netweather                15.46c    Anomaly      1.77c 

Peasedown St John   14.76c    Anomaly     0.8c.    

Thinking of CET in September for a single day, I wonder if there might be a 21C + day early next week.

The only few of these prestigious days I can find since 1940 are 14.9.2016 (21.3); 21.9.2006 and 2.9.1961(21.4) and 5.9.1949 (22.4)

Earlier ones include (and I didn't check further back in detail but knew of the possibilities here) 21.8 and 22.6 on 1. and 2.9.1906 respectively.

Interestingly most of these years had warm or very warm Septembers, with 3 of 16C +; the one that didn't also had the highest day CET (1906).

This will be hard to do mind- if you think, a 25C max day and 15C min day only give 20, although 2011 reminds us how even right at the end of the month the max can be higher.

Interesting to see what happens.

Add: the 1949 day was blessed by an incredible 18.4 min CET on 5th.

After a cool start the first half of September looks very mild. Still close to 17C by the 18th.

The CET could exceed 20C on 7 and 8 September. At the moment it looks like it will be slightly shy of the date record but it could be close

Date         CET est        CET record mean

7 Sept      20.5C           21.1C  1868

8 Sept      20.8C           21.3C  1898

Chart 1

Chart 2

Here is the table of predictions for September

I am going to impose some penalties for late predictions this month. There was a significant shift in the models shortly after the start of the month which means those making very late predictions potentially gained a significant advantage. There are no penalties for those who posted their predictions up to and including 1 September (12 people sent their predictions on the first). 

Those who sent a prediction on the second are going to be penalised by 0.2C (which is Windy Willow, Saint Snow and Duncan McAlister).

Whether Idle sent a prediction at 2236 on the 3rd and will take a 0.4C penalty.

Table

Thanks GW! 

Gosh! Imagine if we had a 17C CET September! 

Given September 2006 came out at 16.8 a 17C CET September is certainly possible? 

Met Office Hadley       15.3c       Anomaly       0.5c provisional to 3rd

Metcheck                    14.77c     Anomaly       1.05c

Netweather                 15.66c     Anomaly       1.97c  

Peasedown St John    15.0c      Anomaly       0.72c. 

At this point it looks very much like we’ll get a high month but then we all know how quickly things can change. 

Thats fair enough - youre the boss.  Mine was down to insane work scedules but hey ho.

Mine was due to having a full message box that wouldn't allow me to send, then forgetting to clear a couple 'later' as I intended (and only just remembered on the 2nd!)

Not really a big issue, though, and I can see your reasoning. 

After watching Gav`s 10-14 day forecast its in the relms of possibility we could have the warmest September on record this year. It looks like all being dependant on what happens with hurricane Larry.Very uncertain after this time next week with some models going for a marked decent into autumn instead. Nevertheless this coming week looks like being one of the warmest weeks of the summer.

Ive sat on the fence with a near average prediction of 14.25c but fully expect this to be way too low.

Not out of the question although it would have to work hard in the second half to match 2006. That year to 18th averaged just under 17.1C; GW has mentioned that this year looks as if it may be around 17 at 18th. Apart from 21st in 2006 (which was very warm) and 24th, the other days were below 17. I'll watch Gav's again with this in mind I think.

Met Office Hadley       15.5c        Anomaly       0.7c provisional to 4th

Metcheck                    14.87c      Anomaly       1.15c

Netweather                 15.69c      Anomaly       2.0c

Peasedown St John     15.2c      Anomaly       0.92c.    

Met Office Hadley        16.0c      Anomaly       1.3c  provisional to 5th

Metcheck                     15.31c    Anomaly        1.58c

Netweather                  16.08c    Anomaly        2.39c

Peasdown St John     16.1c        Anomaly      1.9c.   

Been quite the run of days for CET projections. For a while there were some bonkers GFS runs giving me estimates of more than 18.0°C for the middle days of the month. Not for the daily returns, the mean-to-date! Would be record high for that point in the month.

Anyway, since then GFS has freshened up the weekend and has only intermittent very warm weather thereafter, so CET estimates aren't so eye-watering. They're still notably high though, owing to the impressive very warm / hot spell we're currently in and a lack of sustained switch to a cool pattern in most runs.

ECM was going for warmer but more unsettled outcomes than GFS yesterday but has moved close to it today. Sunday morning could feel decidedly cool as some polar maritime air moves though on a light breeze.

It soon warms up the following week, though. Generally I'm not seeing anything that would force a sustainably cool or very cool pattern to establish within the next fortnight. I'm starting to suspect I may have estimated too low for this month - but then again, I thought that about halfway through last month and...!  

Same here. I deliberately went bang on average as I thought it was going to be a non-descript month. There can have been few months when my prediction has gone so badly wrong at such an early stage in the month. Of course there could still be a very cold final 3rd of the month that would pull the figure back down, but how likely is that, really?

Met Office Hadley      16.7c     Anomaly      2.0c  provisional to 6th

Metcheck                   15.92c   Anomaly      2.19c

Netweather                16.69c   Anomaly      3.0c

Peasedown St John    16.9c    Anomaly     2.62c.   

Met Office Hadley       17.4c      Anomaly     2.7c

Metcheck                    16.50c    Anomaly     2.8c

Netweather                 17.36c    Anomaly     3.66c

Peasedown St John    17.51c    Anomaly     3.23c. 

Met Office Hadley         18.0c       Anomaly      3.3c provisional to 8th

Metcheck                      17.07c     Anomaly      3.34c

Netweather                   17.86c     Anomaly      4.17c

Peasedown St John      17.48c     Anomaly      3.2c.   

I've popped through some of the hot early Septembers and I can't see one since 1868 with a hotter first 8 days- I'm sure someone else can spot one- mean daily CET that is. As we saw early in the month it can change at the drop of a hat though.