Yearly CET

[The PIT]

We're now down to 2.4C per day to avoid a new record. We've already used more than half the degree days "available" this month, just 7 days in. That record could easily have gone earlier than the 18th; I'd say it's now drifting back towards the 17th. A yearly value of 10.7-10.8 certainly looks the most likely now.

Yes it's going to take a sudden drastic change and theres no sign of that at the present time. So record year in the bag it just needs someone to tie it up and take it away.

[Optimus Prime]

What makes this year so remarkable and like no other is that despite it probably will be the warmer year on record we had;

Average January (+0.1c)

Slightly colder than average February (-0.5c)

Colder than average March (-1.4)

Average April (+0.5c)

Average August (-0.1c)

[The PIT]

What makes this year so remarkable and like no other is that despite it probably will be the warmer year on record we had;

Average January (+0.1c)

Slightly colder than average February (-0.5c)

Colder than average March (-1.4)

Average April (+0.5c)

Average August (-0.1c)

The good thing about it we had some below average months. The bad thing we've had some stupendously warm months to over compensate.

[Stratos Ferric]

Pardon my pedantry , but...

I didn't think it was actually possible to declare that the record had been broken until the end of the year. Of course, it's looking more than likely that this will indeed happen, but I'm not sure it's possible to declare meaningfully that the record will be broken on a particular date (except, of course, December 31)

Technically you're right, but that's like saying a football match isn't won until the final whistle despite the fact your team is 22-0 down with about 4 minutes to play. Realistically we are not likely to get many, or probably any, freezing days; on that basis, once all the degree days have been consumed, and given a projection for temps to stay positive, then I think it would be fairly safe to declare a record. The 10.7-10.8 is a mere extrapolation.

[Stratos Ferric]

From here on in we need 8.6C to make 11.00 for the year. It's unlikely, though still quite possible; what is bizarre is that we can even discuss this without danger of being sectioned!

Something above 10.9 for the year is now about evens.

Edited December 12, 2006 by Stratos Ferric

[paul tall]

How many previous record warm (cet) years have we had since say 1988?

[Stratos Ferric]

How many previous record warm (cet) years have we had since say 1988?

Just one (1990), though 1999 equalled it, nudging the previous record (1949) up by 0.01. I have to say I do ponder the maths and potential rounding errors, but I will assume that the annual value has been derived from the average of the daily values to one d.p. rather than the average of the twelve monthly values, which would be subject to more rounding error than the 0.01 difference in apparent values.

[Weather-history]

The warmest 12 month period is November 1994 to October 1995 with 11.07, it could even beat this and be not only the warmest calendar year but the warmest any 12 month period so to speak, although doesn't look likely but its just on the edge of possibilty.

However with May to November 2006 CET average very high, it could fall to May 2006-April 2007. Need to keep an eye on this

Edited December 13, 2006 by Mr_Data

[paul tall]

Just one (1990), though 1999 equalled it, nudging the previous record (1949) up by 0.01. I have to say I do ponder the maths and potential rounding errors, but I will assume that the annual value has been derived from the average of the daily values to one d.p. rather than the average of the twelve monthly values, which would be subject to more rounding error than the 0.01 difference in apparent values.

So its not as if the CET record is being usurped year upon year is it?

For all we know, this could be as good as it gets, a record never to be broken again.

Although admittedly, the last two decades warming trend would point to the chances of it going even higher to be quite a good bet.

[Stratos Ferric]

So its not as if the CET record is being usurped year upon year is it?

For all we know, this could be as good as it gets, a record never to be broken again.

Although admittedly, the last two decades warming trend would point to the chances of it going even higher to be quite a good bet.

On one level that's fair, but another it's a statitstically naive view. For most of the CET's 350 years the temperatures bounced in a narrow band, and the twenty-five year series stayed within a band of about 0.5C right from 1720 to the 1980s. Since 1990 that series has risen by 0.5C out of the top of that band (and is still rising). Ordinarily, in a flat series, with increasing numbers of years, records would be set less frequently - that's simple maths. The last twelve years include, by rank, the 1st, 3rd, 5th, 6th, 7th, 8th and 10th warmest. Over half the last twelve years have been in the warmest 35th. Statistically, one of them might have been expected to be.

Given this it is likely that the record will now be broken repeatedly over the next few decades. Not every year, but I would reckon quite possibly two or three times per decade, if we continue to see the sort of accelerated warming we seem to be experiencing now.

[paul tall]

On one level that's fair, but another it's a statitstically naive view. For most of the CET's 350 years the temperatures bounced in a narrow band, and the twenty-five year series stayed within a band of about 0.5C right from 1720 to the 1980s. Since 1990 that series has risen by 0.5C out of the top of that band (and is still rising). Ordinarily, in a flat series, with increasing numbers of years, records would be set less frequently - that's simple maths. The last twelve years include, by rank, the 1st, 3rd, 5th, 6th, 7th, 8th and 10th warmest. Over half the last twelve years have been in the warmest 35th. Statistically, one of them might have been expected to be.

Given this it is likely that the record will now be broken repeatedly over the next few decades. Not every year, but I would reckon quite possibly two or three times per decade, if we continue to see the sort of accelerated warming we seem to be experiencing now.

I quite liked having an average daytime max of 25ºC in July. The normal average is 19.8ºC for my area. I suppose if the warming continues it will mean more hot summers. Every cloud and all that.

But, its still a big "if"

[Stratos Ferric]

The UKMO have provisionally called what several of us on here have been touting for at least a couple of months now. 2006 is going into the books as setting a new high water mark (quite literally just now) for CET. Reading back through this thread from page 1 makes very amusing reading. Hats off to Reef for a very early spot of potential when all around were doubting / hoping for something cooler. Hats off also to SB for the highest number of reforecasts in a six month period without any of them coming close to nailing the final figure!

If we get a mild winter 11C probably has about even chance of being beaten next year. It's certainly the case, short of a low latitude volcano rearranging itself into the stratosphere, that the glow cast by 11C over the horizon is such that we can be certain it is going to pop its head above the horizon very soon.

[MAF]

have just seen this on BBC site

http://news.bbc.co.uk/1/hi/sci/tech/6177663.stm

[Stratos Ferric]

have just seen this on BBC site

http://news.bbc.co.uk/1/hi/sci/tech/6177663.stm

I don't know what AFF and BFTP will make of such hyperbole laden language. Hard to disagree though; "extraordinary".

[The PIT]

Yes looking like the record is going to be set. Whats the daily temp needed now to prevent it. Something very low.

[Stratos Ferric]

Yes looking like the record is going to be set. Whats the daily temp needed now to prevent it. Something very low.

...as of Monday, 1.6. Less than that now, obviously, given that the minimum last night was (yet again - a recurring theme this autumn) higher than the daytime max ought to have been for this time of year.

[Stargazer]

According to my quick calculation it ain't all over yet (I guess I must be wrong, so would appreciate correction if that's the case). Assuming that we have an average of 10.84 up to Monday (day 345 of 365), an average of 7.01 (still above normal for December) over the remaining 20 days will bring the overall average for the year down to 10.63 (the existing record according to the BBC article). Less than that and we'll miss the record. My calculation is a simple weighted mean: (345*10.84 + 20*7.01)/365 = 10.63. Any comments?

[Stratos Ferric]

According to my quick calculation it ain't all over yet (I guess I must be wrong, so would appreciate correction if that's the case). Assuming that we have an average of 10.84 up to Monday (day 345 of 365), an average of 7.01 (still above normal for December) over the remaining 20 days will bring the overall average for the year down to 10.63 (the existing record according to the BBC article). Less than that and we'll miss the record. My calculation is a simple weighted mean: (345*10.84 + 20*7.01)/365 = 10.63. Any comments?

I haven't checked the BBC values, though I must admit when I heard them my reaction was "there's something wrong there". My calculations are based on year to date, i.e. the situation to the end of November plus December added in. It's possible, as I use the monthly averages, and don't weight for length of month, that there's a rounding error, doubly so if the monthly values are +/- 0.05 and all have gone in one direction; however, all of that together should not make more than 0.1C difference. I'm quite happy my sums, given the above assumptions are correct.

I suspect the error in your calculation must be the starting value for the year to date and / or where that has been measured up to. I have 11.22 to the end of November. Adding in 8.3 to the 11th December gives a new average YTD of 11.13. That starting point makes it a whole lot harder to get down to 10.63. Where does 10.84 come from?

According to my quick calculation it ain't all over yet (I guess I must be wrong, so would appreciate correction if that's the case). Assuming that we have an average of 10.84 up to Monday (day 345 of 365), an average of 7.01 (still above normal for December) over the remaining 20 days will bring the overall average for the year down to 10.63 (the existing record according to the BBC article). Less than that and we'll miss the record. My calculation is a simple weighted mean: (345*10.84 + 20*7.01)/365 = 10.63. Any comments?

Having checked the BBC site, and double checked my values for the months from Philip Eden's site (they're all there or thereabouts) I'm wondering whether what the BBC is not making clear is that the 10.86 (or whatever) that they're citing is actually a provisional value to the year end as of December 13th, as opposed to an actual average to December 13th. That latter would definitely stack up with my own projection and would make most sense. In that case yet another example of sloppy journalistic english - the bane of my life is journalists who have learned media skills but either know diddly squat about the subject they're writing about, and / or have no analytical skills.

[West is Best]

I must admit - the moment I saw the 10.84C I thought 'er, no actually'. No idea where that figure comes from, but it certainly isn't based on the official CET from 1st January to now.

[Stargazer]

The BBC web page quotes "The Central England Temperature Record (CET) is the oldest continuous dataset for temperature anywhere in the world...The average temperature for the year up to 13 December stands at 10.84C. In the 1950s, the CET showed an average of about 9.4C." On the surface this looks like a clear, definitive statement that the CET for the year to date is 10.84 deg C . I guess I shouldn't believe everything I read

[Stratos Ferric]

The BBC web page quotes "The Central England Temperature Record (CET) is the oldest continuous dataset for temperature anywhere in the world...The average temperature for the year up to 13 December stands at 10.84C. In the 1950s, the CET showed an average of about 9.4C." On the surface this looks like a clear, definitive statement that the CET for the year to date is 10.84 deg C . I guess I shouldn't believe everything I read :unsure:

Certainly sloppy English, but based on what was said on the news last night there's no doubt that my suggestion re extrapolation is the answer. I think on one of the bulletins I saw the statement made was to the effect of 'yearly [CET] is expected to come in around 10.8'.

[Weather-history]

Simon Keeling has made a big boo boo in his article on Weatheronline, saying it might not be a record warm year

http://www.weatheronline.co.uk/feature/2006/12/14_sk.htm

He has taken Philip Eden's Manley CET values for November and December as shown as actual values when infact they are actually 1971-2000 averages for those months.

http://www.climate-uk.com/provisional.htm

Yes, if November and December were average, then 2006 would only equal 1990 in the Manley series but November was about 8.0 and not 6.9, so its higher than Simon is making out.

Edited December 15, 2006 by Mr_Data

[Stargazer]

Just taken a look at Philip Eden's page. The November and December figures are shown in red (all the rest are in black) - from that I think we can infer that he hasn't updated the November CET figure on the page yet.

According to my calculations we are running at a little over 11.1 up to December 15 (give or take rounding errors). My interpretation of the BBC article was evidently incorrect (although if I were being pedantic I would claim that the article itself was incorrect and my literal interpretation was fine).

Edited December 15, 2006 by Stargazer

[Stratos Ferric]

Just taken a look at Philip Eden's page. The November and December figures are shown in red (all the rest are in black) - from that I think we can infer that he hasn't updated the November CET figure on the page yet.

According to my calculations we are running at a little over 11.1 up to December 15 (give or take rounding errors). My interpretation of the BBC article was evidently incorrect (although if I were being pedantic I would claim that the article itself was incorrect and my literal interpretation was fine).

You shouldn't be apologising. As I commented, sloppy English on their part: there's no doubt that as written on the BBC site the inference was 10.8x up to present.

[Optimus Prime]

The actual yearly CET so far is exactly 11.0166666666

If the December averages;

7.0 the yearly CET will be 10.8833333333c

6.0 the yearly CET will be 10.8c

5.0c the yearly CET will be 10.71c

4.0c the yearly CET will be 10.63c

3.0c the yearly CET will be 10.55c

2.0c the yearly CET will be 10.46c

1.0c the yearly CET will be 10.38c

0c the yearly CET will be 10.3c

-1.0 the yearly CET will be 10.21c

-2.0 the yearly CET will be 10.13c

....

-10c the yearly CET will come in at average (9.46)

....

-50 the yearly CET will be 6.1c

....

-100 the yearly CET will be 1.9c

It will literally take the universe to freeze for us to get an average yearly temperature....unlikely.

Edited December 15, 2006 by Optimus Prime