I would find it fascinating if someone drew up a 'potential absolute maximum' for the UK perhaps using interpolation of some of the highest date records. 

For example, if 1st Nov can get 22 (in 2015)  and 1st Oct (in 2011) get 30, then half way between (ie about 16th Oct) 26 should be possible.

Just a thought. Of course the true potential absolute max could well be higher, assuming those particular date records are not true absolutes.

I recognise of course that there might not be a truly linear pattern, and also that warming may affect these figures over time. 

But based on current achievements, I wonder what's possible.

Some very extreme peaks include:

23.9°C on 9th March 1948,

29.5°C on 16th April 1949,

34.4°C on 3rd June 1947,

38.5° on 10th August 2003,

30.6° on 27th September 1895,

23.6°C on 31st October 2014

20.0°C on 20th November 1906.

One can't help but feel that that May, and particularly July, have not reached their full potential. I reckon almost 34 is possible at the end of May and above 39 in late July, from these peaks. 

Almost got to 20C in the middle of February

13th February 1998: 19.7C

There must be a chance of a 20C being recorded sometime during a winter month, for instance, if that exceptional mild air mass of mid February 1998 had occurred during the last week of February 1998 instead. 

There was a 20C recorded on the 2nd March 1977.

Good points Kev; 13th Feb 1998 should definitely be added to the 'peaks' list.

Whilst 2.3.77 was impressive for so early in the meteorological spring, with not far short of 20 on 13th Feb and 23.6 on 9th March, I reckon that early March 20 is an under-achievement. What do you think?Edit: 25.9 on 18th October in 1997 is also a shouter.

I reckon 34.4c set at Gravesend on 13th September 2016 reached absolute potential maxima for that date. 

That 29.9c set at Gravesend on 1st October 2011 could have gone a smidgen higher.

I also maintain that the absolute maxima of 38.5c set way back in 2003 is a record there for the taking with some style.

I think I'm with you on that Steve.

I think the 9th March 23.9, if accurate, is top drawer like 13th Sep: just a week and a half out of winter, and only a degree lower than the month record, so early on. 

29.5 on 16th April is another that must be close to the top. On these days, everything must have come together right.

Perhaps, if someone has the data and time available, one could draw a graph of successive record temps (y axis) against year (x axis).

You would probably see a graph steadily levelling off, reaching the maximum asymptotically

Already on the case. Here it is.

TABLE

From that we can conclude that 21.0C is possible in february.

If my memory serves me rightly imby we had a 21c+ around March 12th 2011 and a 27c+ at easter that same year. Also we had an 18c+ in late Feb 2012 after seeing overnight mins down to -11c earlier that month.Also there was Dec2015 with some places as high as 17c 

I certainly think 40c is possible with the perfect set up at some point in the last 10 days of July.Given how good this summer was the maxes were a tad disappointing and expected a few days in the high 30`s in some spots in the south.

Maybe next year.

What I have always found surprising about the date records is that even after well over 100 years of observations there is still so much day to day variation. Somehow I would have expected the line to be a lot smoother after that time period.

That 13th of June is a strange one.

Might be worth looking at the CET daily maxima for the 12th, 13th and 14th June since 1878 and getting an average to see if there is a definite drop on the 13th June.

Fascinating graph, though not what I had in mind.

Also interesting if trivial to look at minima; 7 Jun and 22 Sep (if I can read the axis correctly at this scale) have notably missed out on a decent date record.

Edit: Oops! transposed a week's column - should be 13 Jun - see subsequent posts below

Likewise Sept 22nd. A genuinely cooler day or just statistical variation? After all with 365 days in the year you might expect one or two just by chance to have recorded a max temp notably lower than surrounding days.

Just had a look at this and in fact 13 June has the highest mean of those three days.

Average Max CET 1878-2018

12 June 18.48C

13 June 18.57C

14 June 18.40C

If instead you take a mean from 1950-2018, to reflect the fact that it has been warmer in more recent years and hence more likely the date record would be broken, you get the following

12 June 18.42C

13 June 18.78C

14 June 18.80C

The 12th is noticeably cooler than the other days but not the 13th.

Same here. No significant drop in the overall mean maximum compared to surrounding days

1878-2018

21 Sept 16.60C

22 Sept 16.57C

23 Sept 16.49C

1950-2018

21 Sept 17.32C

22 Sept 17.22C

23 Sept 17.24C

17th june looks a bit pants doesnt it

That's the 13th June. It's the only day in the meteorological summer that has weirdly never recorded 30C.

what - like anywhere in the UK? Ever (within recorded times)

Thats got to be a bit of a miscalc on someones behalf.  

Nope, there has been no official 30C recorded on that date anywhere in the UK.

The highest is 28.3C

It's just a statistical fluke.  It has already been shown that the CET for that day isn't significantly lower than the immeditely surrounding days. So it's simply mere chance that date has never landed within a significant heatwave.

Has someone got a link to the record max temp data for each day?

http://torro.org.uk/hightempsyear.php 

The max for the 10th is 25.6C, BBC going for 24C somewhere tomorrow. So it's certainly possible that may be beaten but it's probably out of reach.