May CET

[Snowly does it]

Being wrong is something I am used to - it only in fact rose to 12.5 yesterday (thanks to Phillip Eden for this)

[Anti-Mild]

12 looks the target now.

Which would be a little cooler than last May, and the last time a month was cooler than it's previous incarnation was last October.

[summer blizzard]

Assuming no further rise, and a 0.4C drop from now until the end of the month, i think 12C-12.2C is likely depending on roundings, which would make it average for the past ten years.

[Bottesford]

And the spring record? I forget what May needs for us to get the warmest spring on record...

[Snowly does it]

Which would be a little cooler than last May, and the last time a month was cooler than it's previous incarnation was last October.

Yes and sub 12.1 would be the first month under the 10 year mean since August. Conversely, 12.2 would equal the spring record. Either side of 12.1 is the critical figure this month for records, trends, interest and errata!

Not enough under last May to turn the tide of a record obliterating year, we would need to see a run of '30 year average' months to draw it back from the brink of popping the elevenses balloon.

[Duncan McAlister]

And the spring record? I forget what May needs for us to get the warmest spring on record...

Spring record (1893) = 10.20

A May of 12.2 would give an exact Spring CET of 10.19, so 12.3 would be required, which would give us 10.22.

[Optimus Prime]

The spring record doesn't look likely it will fall. But it should top the top 3 or 4 of warmest Springs ever.

[Snowly does it]

Spring record (1893) = 10.20

A May of 12.2 would give an exact Spring CET of 10.19, so 12.3 would be required, which would give us 10.22.

??

7.2 March plus 11.2 April plus 12.2 May = 30.6 divided by 3 = 10.2, record equalled

12.3 to take it outright?

[Don]

Certainly very close to the Spring record and the 4th season in a row to be exceptionally above average! How long will it continue?

[Duncan McAlister]

??

7.2 March plus 11.2 April plus 12.2 May = 30.6 divided by 3 = 10.2, record equalled

12.3 to take it outright?

You're not taking into account April is only 30 days. True enough I'm nitpicking, normally it wouldn't really matter but in this situation this tiny margin makes the crucial difference.

7.2*31=223.2

+

11.2*30=336

+

12.2*31=378.6

/

92 days= 10.19

Edit: No you're right, my apologies! Just checked this winter's average and it seems the Met Office calculate it your much quicker way!

So the record still on :wub: ....

Edited May 26, 2007 by Duncan McAlister

[The PIT]

You're not taking into account April is only 30 days. True enough I'm nitpicking, normally it wouldn't really matter but in this situation this tiny margin makes the crucial difference.

7.2*31=223.2

+

11.2*30=336

+

12.2*31=378.6

/

92 days= 10.19

Eh???? If you take the months as a whole the number of days doesn't come into it. So Snowmaiden is correct.

[Duncan McAlister]

Eh???? If you take the months as a whole the number of days doesn't come into it. So Snowmaiden is correct.

I stand corrected!

Edited May 26, 2007 by Duncan McAlister

[Snowly does it]

No problem Duncan!

As for the record - an interesting one this, the record could fall with March 36th and May 62nd in their all time lists. In other words, the record would be entirely due to an extraordinary April, the other seasonal months being warm but not excessive or indeed unusual and within a degree of the long term average.

I think May will edge in a little below the record as I see no rises from here until possibly the last day and it could dip below 12 before then. It would need to drop to 11.7 to fall outside the top 100 and be the first month since August 2006 to not be in the top 100 for that month.

Edited May 26, 2007 by snowmaiden

[Duncan McAlister]

But it would still be 10.2

If you are only quoting your daily temperature to one decimal place, then you should only quote the average to one decimal place too. At least that's what we were taught at school in the 70's.

I've seen this argument on here before and agree really, but the Met Office practice is to give monthly averages to one decimal place and seasonal/annual averages to two.

http://hadobs.metoffice.com/hadcet/ssn_HadCET_mean_sort.txt

[Iceberg]

Given the fact the GFS has undercooked temps today by 1 or 2 C (it's max a 9.00am for me was 8, in real life it was 10C.)

Today and Monday will be a bit down (maybe dragging it down by .2) Wed pretty neutral maybe taking up by 0.05, thursday looks to be warm again with 20C being reached taking up by 0.1C, so in total only dropping very slightly down from yesterdays figure. With the above I think a record breaking spring will just be hit, thanks to the final may figure.

[West is Best]

Yes and sub 12.1 would be the first month under the 10 year mean since August. Conversely, 12.2 would equal the spring record. Either side of 12.1 is the critical figure this month for records, trends, interest and errata!

As for the record - an interesting one this, the record could fall with March 36th and May 62nd in their all time lists. In other words, the record would be entirely due to an extraordinary April, the other seasonal months being warm but not excessive or indeed unusual and within a degree of the long term average.

I think May will edge in a little below the record as I see no rises from here until possibly the last day and it could dip below 12 before then. It would need to drop to 11.7 to fall outside the top 100 and be the first month since August 2006 to not be in the top 100 for that month.

Hi Snowmaiden,

One of the reasons I griped about your new-fangled '10 year mean' is highlighted by your comments. You are chopping and changing the mean to suit. One minute you are measuring against the long-term average, the next against your ten year one. It's this sort of statistical 'flexibility' (to be euphemistic and polite) that brings statistics into disrepute. Having said that, there are many ways to skin a cat. For instance, your '36th and 62nd' for March and May actually masks the fact that all 3 months this spring will have been substantially above average.

The 3 means which most metereologists consider significant are the 1961-1990, 1971-2000 and then the longer term 100 year rolling average. The figures for these in turn are:

March 5.7 6.3 5.9

April 7.9 8.1 8.1

May 11.2 11.3 11.4

Against these means, the current spring of 7.2 11.2 and ?12.4 looks pretty stonkingly warm right across the trimestre.

Incidentally, a word of caution to those of you expecting a major correction (I notice SB continually predicting a fall of 0.4C for instance): don't forget that most corrective capability for May occurs early in the month thanks to the frost potential. Towards the end of the month even with grim and grey cold days like today you are unlikely to see the major downturn in average temperature that I think some are expecting.

[Snowly does it]

One of the reasons I griped about your new-fangled '10 year mean' is highlighted by your comments. You are chopping and changing the mean to suit. One minute you are measuring against the long-term average, the next against your ten year one. It's this sort of statistical 'flexibility' (to be euphemistic and polite) that brings statistics into disrepute. Having said that, there are many ways to skin a cat. For instance, your '36th and 62nd' for March and May actually masks the fact that all 3 months this spring will have been substantially above average.

Just illustrating a point or two Richard.

I'm not chopping and changing to suit, there are just different arguments. There is the argument of 'warming' - for that we should refer to the 71-00 mean. Then there is the argument of the 'lurch' or Calrissian's 'jump' - for that the shorter term average is more useful.

'36th and 62nd' - show that whilst warm, whislt above average, neither March nor May is unusually warm or unusually above average - it is April and April alone that stands out for this Spring. March and May are round about where I'd expect them to be.

Frosts are possible Monday and Tuesday night and daytimes are not looking warm, I do expect a corrrection and I think 12.4 is incredibly optimistic - it would require 12 per day every day (including yesterday). 11.9 to 12.1 for me and no record.

[Anti-Mild]

Whilst I agree that 12.4c is a little on the high side for the end of month CET (P.Eden has it at 12.4 up to the 26th), I also think that 12.1c is a touch low.

The final 5 days of the month will, I think, drop the CET to around 12.2c according to Climate UK. Then comes the wait for the official word from the MetO to see whether the record is broken.

For what little it is worth I expect the record to remain standing, but only just!

[Weather-history]

As with last year, the interest may turn away from the CET and towards how wet this May turns out. It has happened only once before of two consecutive Mays with a rainfall average of 100mm+ (1796, 1797). We'll have to see what happens.

It looks as though we are going to breech the 100mm line. If it does and is confirmed by Hadley, it will be in the top 25 wettest Mays on record

Edited May 27, 2007 by Mr_Data

[West is Best]

it is April and April alone that stands out for this Spring. March and May are round about where I'd expect them to be.

No I think this is incorrect. For a 3 month period to break a record requires more than one 'blip' month, and I think you are missing the point here. To have a remarkably warm month sanwiched between two other months that are also well above average is the truly remarkable point. We expect 'spikes' to occur, but these are usually offset by corrections. This spring is remarkable not merely for April, but because there has been no correction to that anomalously warm month.

As for what you expect, that's the problem. If I set an arbitrary benchmark of expectation, which I may as well pull out of the hat willy nilly, then I can come up with whatever fanciful interpretation I like and post it as spurious backing for something less-than-remarkable.

Edited May 27, 2007 by West is Best

[Nick H]

As for what you expect, that's the problem. If I set an arbitrary benchmark of expectation, which I may as well pull out of the hat willy nilly, then I can come up with whatever fanciful interpretation I like and post it as spurious backing for something less-than-remarkable.

Good point. The fact is, we're so used to these +1C above the mean months that people fail to recognise that such a sequence, in itself, is remarkable.

However, I would be surprise if we took the spring record from here. I can't see a maxima of 12C from now on in, let alone average of 12C.

Edited May 27, 2007 by Nick H

[Weather-history]

"all 3 months this spring will have been substantially above average"

Depends how you define substantially above average.

Is March 2007 substantially above average?

I think that is Tamara's point, Richard.

One thing though, Richard, March and April 2007 is the warmest March-April combination on record but infact March didn't contribute much to breaking this record.

If you look at the old record of 1961, (March: 8.2, April: 10.0). You can see they both made sizeable contributions to breaking the record.

Based on that, I can see where Tamara is coming from.

Edited May 27, 2007 by Mr_Data

[Nick H]

"all 3 months this spring will have been substantially above average"

Depends how you define substantially above average.

Is March 2007 substantially above average?

I think that is Tamara's point, Richard.

From the Met Office:

(March 2007 was) generally around 1.5 to 2 °C above average across England

They used to categorise it, i.e. "well above average", "exceptionally above average" etc but not anymore. Still I would imagine March 2007 would fall into the latter category.

36nd doesn't sound that impressive, but if you do the maths, there are 349 years of CET data, that's almost in the top 10% of warmest Marches on record (10.3%). In other words, Mar 2007 has a return period of 10 years. I think that qualifies for the "exceptionally above average" category.

Even at 62nd, May 2007 will fall well within the warmest 20% of Mays on record. I would have thought that would classify as "well above average".

So, take the 3 months together and you have "exceptionally above average", "warmest on record" and "well above average". It's not surprising that we're heading for the 2nd/3rd warmest spring on record.

Edited May 27, 2007 by Nick H

[Snowly does it]

No I think this is incorrect. For a 3 month period to break a record requires more than one 'blip' month, and I think you are missing the point here. To have a remarkably warm month sanwiched between two other months that are also well above average is the truly remarkable point. We expect 'spikes' to occur, but these are usually offset by corrections. This spring is remarkable not merely for April, but because there has been no correction to that anomalously warm month.

As for what you expect, that's the problem. If I set an arbitrary benchmark of expectation, which I may as well pull out of the hat willy nilly, then I can come up with whatever fanciful interpretation I like and post it as spurious backing for something less-than-remarkable.

In an established warming trend (I take it we all accept that the world is warming?) where the Uk has demonstrably warmed by somewhere in the region of three quarters of a degree to a degree then I hardly see anything significant or unusual about two months which come in approximately three quarters of a degree to a degree above the 'historic' average.

Fine though, so as not to detract from the already established facts of Global Warming I retract my lack of surprise at two months coming in at almost precisely the level of warming observed and am now astounded that March and May have come in slightly less than a degree above the 1971-200 mean. I have in fact spluttered coffee all over my keyboard in shock

How long before 71-00 is ditched as totally useless in our changed reality?

[summer blizzard]

I agree with Tamara in that the record Sring is almost solely due to April, this is because when the March and May anomolies are added together, they are still over 1C short of the April anomoly, infact, if we add the anomolies together assuming May to have an anomoly of 0.8C, and calculating percentages, April was responsible for 60%-70% of this springs anomolous warmth.