The important word there is "weighted", and the number of days in each month is what determines the weightings. Imagine a very strange calendar where June had 90 days and July and August had only one day each. (So still 92 days in total.) Hopefully it's obvious that, with that calendar, if June had a mean CET of 20.0C and July and August each had a mean CET of 15.0C, the mean CET for the summer would be only a smidgeon under 20 and certainly not (20+15+15)/3. Each day's temperature has to make an equal contribution to getting the mean CET for the season, regardless of the length of the month that it's part of. To ensure that, you multiple each month's mean CET by the number of days in the month, to give what you might call "degree-days". Then you add the three monthly values and divide by the total number of days in the season. 

If only my maths teacher had explained things that way, I might have done better than a grade C in O level!  

Months are simply artificial groupings of days.

If you want to work out the average temperature for a group of days, say 1st June - 31st August, add up the temperature for each day and divide by 92. 

Yes, that is just a different way of doing the same calculation.

Annual CET competition - July update

redmoons was the closest in July with an error of 0.02C closely followed by our competition leader Darren S with 0.03C and Dingle Rob and Frank H also with 0.03C errors.

Here is the July table. No change at the top. Darren S still comfortably ahead of Stormchaser and Gusty. Will we see some changes in August?

A star next to a name indicates that person has reached their highest position in the table for 2018.

Link to larger TABLE

July was a hot month, the third warmest July since the CET series began in 1659.

Mean CET of 19.13C, only the fourth time in any month the CET has exceeded 19C.

The maximum CET of 25.4C was second only to the 25.6C recorded in 2006.

The minimum CET of 12.8C was the 14th equal of all time. 

The May to July CET mean of 16.15C is the highest ever beating the 15.95C of 2006.

The April to July CET mean of 14.60C is the highest ever beating the 14.13C of 2006

There were 2 CET date records in July on 26th and 27th.

TABLE1

TABLE2

Here is the summer CET tracker

We are currently 2C above the 1981-2010 mean. In terms of other hot summers we were behind 2006 and 1976 at the end of July.

CHART

Despite the last four months being very mild, the 12 month rolling CET remains low because prior to April we had several months of cold or average temperatures. 

Currently the rolling mean is at 10.38C. If we keep the warm weather going into Autumn expect this to rise well above 10.5C.

Thank you very much GW!  Interesting stats with some records broken showing just what a great summer this is!  

July has been one of the most exciting months I think we’ve had in the competition thread!  Thanks to everyone who has contributed and shared the excitement and well done to Redmoons and all who came close!  

Thank You GW for the monthly stats.. 

The April to July figure is particularly impressive. It's quite something to break a record for a period as long as four months by as much as 0.47C.

Fasinating stats and much appreciated GW.

Just goes to show almost for the past 12 months the uk has gone to a continental type climate from its usual maritime climate,we had a long cold winter followed by a rapid April warm up, a very short spring and a long hot summer.It will be interesting when things return to normal.

I've just scanned through the HadCET max data, and as far as I can see July 2018 is the only calendar month where every single day reached or exceeded 20C. July 2006 and 2003 didn't make it, nor did July or August 1995, 1976  or 1989, nor August 1947 or August 1975. There might be another one in there somewhere earlier on. Another trophy to July 2018.

Here's a statistical nerdy point.

The last time a July returned a CET of XX.4C?

1975 

It is with great regret that I have to announce that the HadCET data set has been further adjusted with 19.7C on 28th meaning that the above comment is no longer true. Why can't they just leave things alone?

Edit: hah! 2018 is still tops- joint second is July 06 and July 83 which both have 19.5, rather than 19.7 as their low. None of the others are in the running.

I think it's only 19.7 on the provisional/estimated data. Still over 20 on the final version

Incidentally July 1934 was also close being same as 2006 and 1983 with 1 day at 19.5C and all the rest 20.0 or more

Thank you- reassurance and additional information. 

Toys back in buggie.

The final figures for July are in. Here is the updated list of maximum temperatures for each day.

We managed an extra day with 30C as Cirencester reached 30.4C on the 22nd.

Also a first ever entry in this list I think for Dersingham Bog in Norfolk. The 33.5C on the 23rd sets a new date record and means we broke 3 date records in July.

TABLE1

We now stand at 27 days with 30C plus this year. Those 27 days fell during a period of just 44 days.

TABLE2