July CET

[West is Best]

I don't think I said it was did I? I was more projecting forwards given that the fundamentals currently favour sustained (relative) cool and given that the second half of June was sub-par.

Ah - ok. Understood.

By the way, weren't the last few days of June above average? Thought it dipped and then rose right at the end? Or am I imagining it?

[Stu_London]

Ah - ok. Understood.

By the way, weren't the last few days of June above average? Thought it dipped and then rose right at the end? Or am I imagining it?

According to climate.uk it was a straight gradual downward line although the daily graphs point to a highish minima on the 30th and the 1st of July was something like 16.7 for Manley and 16.9 for Hadley

[kold weather]

Gavin---I still think the odds strongly favor a below average month, though it may be a touch early to say below 16C I think that too is looking increasingly likely. I relaly can't see much of a dip for this week as the maxes will probably average out about 18-20C with mins probably somewhere between 10-12C and so we may drop a small bit but nothing that will take us sub-15C!

[stormchaser1]

Evening everyone

First i apologise if this has been mentioned before ,but since 1985 i have worked out that there has been

3x 18c = 1989=18.2c

======1994= 18.0c

======1995=18.6c

there has been 6x15c

1986= 15.9c

1987= 15.9c

1993= 15.2c

1998= 15.5c

2000= 15.5c

2004= 15.4c

there has been 7x16c

1985= 16.2c

1990= 16.9c

1992= 16.2c

1996= 16.5c

1997= 16.7c

2002= 16.0c

2005= 16.9c

there has been 4x 17c

1991= 17.3c

1999= 17.7c

2001= 17.2c

2003= 17.6c

and there has been 1x 19c

2006= 19.7c

Overall since 1985 the average temp has been 16.57c

nigel

Edited July 15, 2007 by stormchaser1

[Gavin P]

Cheers Kold. :wacko:

[Stu_London]

Latest from climate-uk.com

July 1st to 15th

CET 15.3C (-0.9C)

Rain 52.9mm (191%)

Sun 72.0hr (74%)

So - looks like the maxima under the rain yesterday were supressed enough to keep the CET on hold. However a warm night and a warmish day today will probably see us rise by a tenth or two tomorrow. After that we go back to the pattern of last week with a daily mean likely between 14 and 15.5. We may slip back to somewhere near 15.1 or 15.2 by the end of next weekend, leaving the fate of the CET in the last quarter of the month - Ensembles would appear to suggest similar conditions prevailing which would leave us floundering a whole 1C below the 71-00 average.

Rainfall is worth keeping an eye on. We will probably pass the total average for July by today or tomorrow meaning yet another wet month (7 from 8 now). Other records to look out for are 100mm being reached (as was the case in May and June) as well as various other records Mr D is tracking in the rainfall thread.

With the outlook as it is, it is difficult to imagine anything other than another disappointingly dull month.

[Gavin P]

How much it drops later in the week will depend on minima and how low it gets. GFS 00z going for some pretty cool nights at the end of the week;

http://85.214.49.20/wz/pics/Rtavn7817.png

http://85.214.49.20/wz/pics/Rtavn10217.png

http://85.214.49.20/wz/pics/Rtavn12617.png

http://85.214.49.20/wz/pics/Rtavn15017.png

Edited July 16, 2007 by Gavin P

[Stu_London]

How much it drops later in the week will depend on minima and how low it gets. GFS 00z going for some pretty cool nights at the end of the week;

http://85.214.49.20/wz/pics/Rtavn7817.png

http://85.214.49.20/wz/pics/Rtavn10217.png

http://85.214.49.20/wz/pics/Rtavn12617.png

http://85.214.49.20/wz/pics/Rtavn15017.png

They are not the most reliable, however if those values were to come of the daily means of around 14C would be the order and we may lose 0.3C over 5 days. We could be 15.2 by next Sunday which with 9 days of the month remaining would require to remainder to be around 18C to reach 16.0C and 19.7C to break average (16.5C)

My narrow spread on the final figure now is 15.2C to 15.7C

Edited July 16, 2007 by Stu_London

[Weather-history]

Philip's data on the 16th Jun-15th Jul period makes woeful reading

14.9 (-0.5)

162.9mm (243%)

129.3hrs (68%)

http://www.climate-uk.com/page2.html

[The PIT]

Philip's data on the 16th Jun-15th Jul period makes woeful reading

14.9 (-0.5)

162.9mm (243%)

129.3hrs (68%)

http://www.climate-uk.com/page2.html

Ah sneaking up towards average temp. Trying to be positive here.

[Stratos Ferric]

They are not the most reliable, however if those values were to come of the daily means of around 14C would be the order and we may lose 0.3C over 5 days. We could be 15.2 by next Sunday which with 9 days of the month remaining would require to remainder to be around 18C to reach 16.0C and 19.7C to break average (16.5C)

My narrow spread on the final figure now is 15.2C to 15.7C

I think a rise of 0.2 to 0.3C is likely over the next three days. Getting back to 15.2 by the end of the w/e would require an unlikely looking scenario, or major changes in the near-time sequence. For that reason I think your spread is looking a little low. With such a volatile set-up and the current projection being cool any error is more likely to be upside.

It's still surprising, despite all the dross, that we're not further below par than we are. Yet again the relatively mild (cloudy) nights are dampening some of the potential impact.

[Bobby]

When was the last "well below average" month? By Met Office classification, think it's 1.5c below. Must have been quite some time ago.

[shuggee]

Oh, I don't know SF - all it takes is a bit of sunshine and the temp shoots up. Brighton yesterday was uncomfortably hot (but there again I am fat and wear a lot of black). The damp, cloudy and warm nights do have the impact you suggest of course.

[Stratos Ferric]

When was the last "well below average" month? By Met Office classification, think it's 1.5c below. Must have been quite some time ago.

Depends on how you define "quite some time", but using my preferred rolling measure as opposed to the standard ten year stepped interval, more recently than I suspect you think. March 2006 came in at -1.6. Before that Dec 96 and Jan 97 both breached the mark. The last time -2.0C was exceeded was October 1992 (-2.7).

Last year (December-November for this purpose), for all it's record breaking headlines, still managed sub-par months, and for all the warming in the modern climatic regime, we still average 2-3 sub-par months a year. What's notable is that the balance to warmth is also matched by the tendency for upside warmth to be bigger than downside cool.

Oh, I don't know SF - all it takes is a bit of sunshine and the temp shoots up. Brighton yesterday was uncomfortably hot (but there again I am fat and wear a lot of black). The damp, cloudy and warm nights do have the impact you suggest of course.

Some of what you felt was probably very high humidity, though as we do have to keep reminding ourselves, and as you say, with the sun very high it doesn't need to be out for long for temperatures to jump fairly smartly in slow moving air. I think that latter has also helped pin temperatures down this summer; I'd hasard that as much as dullness has capped things, wind mixing the boundary has probably been no less a factor. I'm not aware that formal stats are maintained anywhere but I'd be surprised if this summer isn't up there with the breeziest on record.

[Stu_London]

I think a rise of 0.2 to 0.3C is likely over the next three days. Getting back to 15.2 by the end of the w/e would require an unlikely looking scenario, or major changes in the near-time sequence. For that reason I think your spread is looking a little low. With such a volatile set-up and the current projection being cool any error is more likely to be upside.

It's still surprising, despite all the dross, that we're not further below par than we are. Yet again the relatively mild (cloudy) nights are dampening some of the potential impact.

I think we may get a rise of 0.2C today, however with even the BBC website citing widespread single figure minimas for the rest of the week in their monthly outlook it does look like the CET will be heading south after today. If we do hit 15.5C tomorrow it only needs to be 14.4 between then and Saturday to get back to 15.2C

Mins of around 9C and maxes of 19C to 20C are very much on the cards I would say, unless the outlook has changed since 5am this morning

[West is Best]

Last year (December-November for this purpose), for all it's record breaking headlines, still managed sub-par months, and for all the warming in the modern climatic regime, we still average 2-3 sub-par months a year. What's notable is that the balance to warmth is also matched by the tendency for upside warmth to be bigger than downside cool.

I think it's important to point out that this is only the case if you apply entirely arbitrary self-made statistical baselines - baselines which are imo of far too short a time-period to have any meaningful metereological, climatic or statistical worth. They are, in particular, predicated on the assumption that global warming is taking place, and which is having an immediate effect on UK temperatures. Yet, this predication is using statistics to prove a predetermined stance, rather than allowing ourselves to be informed by the statistics. In other words, it's highly assumptive and even manipulative. Climatic variations are properly visible over a sufficiently extended baseline so as not to produce skewed results at best, utterly meaningless ones at worst.

It is important in my view to stick with the official designations here as given by the Hadley Centre and to a lesser extent Philip Eden's Manley readings, rather than some will o' the wisp approach.

The Met Office have produced a lot of very good material on the subject of averages, as official guardians of this, which repay careful reading. http://www.metoffice.gov.uk/climate/uk/about/methods.html

The idea that a month is 'well below' if it surpasses 1C or 1.5C below the average is itself rather haphazard. A safer method of calibration is to take a quintile segmentation: thus the average falls 20% either side of the 30 year mean, with 'above' and 'below' in the two respective bands either side of this (60-80% and 40-20%). 'Well above' and 'well below' fall in the 1/5th and 5/5th segments of the 30-year mean. (The Met Office also use the 1961-1990 stats alongside the 1971-2000 stats)

I haven't got the time just now to check the last occasion a month fell in the bottom 1/5th against the 30 year average and was, thus, 'well below average' but it's easily checked. The 1971-2000 mean is the middle figure here (the first one is the 1961-1990 average, and the third one is the 100 year rolling average).

January 3.8 4.2 3.9

February 3.8 4.2 4.1

March 5.7 6.3 5.9

April 7.9 8.1 8.1

May 11.2 11.3 11.4

June 14.1 14.1 14.2

July 16.1 16.5 16.1

August 15.8 16.2 15.9

Sept 13.6 13.7 13.6

Oct 10.6 10.4 10.2

Nov 6.5 6.9 6.6

Dec 4.6 5.1 4.7

This can then be checked against the official CET here: http://www.metoffice.gov.uk/research/hadle.../HadCET_act.txt

It would all need to be mounted up into a spreadsheet I guess ... something SF is very good at, even if I have a considerable disagreement over his treatment of statistical baselines for working out the averages! (No offence SF - I think you know my views on this!)

By this criteria, which seems much fairer, 'well below' average months ought to be occuring more regularly than with the more arbitrary fixed-figure method.

Edited July 16, 2007 by West is Best

[Stratos Ferric]

WiB, I'm not quite sure what statistical point you're querying there.

I have sympathy for the labelling of "cold", "very cold" etc. to be done on a relative basis, however, a case can alos be made for an absolute scale: after all, what we feel is ABSOLUTE not RELATIVE. Hence, 12C in January feels very different to our senses compared with 12C in the middle of the day in July. I think that in this regard I agree with the UKMO. Furthermore, in comparing across climatic regimes, we'd get all sorts of nonsense. On the equator, where variation over any timescale in units of 24h is fairly slight, adopting the use of standard deviations from the mean would have a completely different interval between grades compared with more marginal temperate climates like our own.

Given that our climate is always changing, then so long as a decent rolling average is maintained, the adoption of intervals of, say, 1C either side of this mark to attribute qualitative labels is, to my mind, reasonable utility. A relative scale would have different intervals between each grade (such is the effect of the normal distribution) and also between months, so whilst a case might be made from a purely pure statistical viewpoint, it would cause mass confusion in the public at large, and would, in any case, be subject to constant review.

As Mr Holmes and I are frequently at pains to point out, the important thing is to state clearly your assumptions ahead of any numerical assertions, thereafter - within reason - it's a matter of horses for courses.

Re my use of the rolling average. It is statistically a much better baseline. I would have more sympathy for UKMO if they stepped the baseline forwards immediately each ten year chunk had been completed (alowing a little time to check data, etc.). I have a suspicion that this is just an anachronism dating back to a time when calculations were more time consuming. After all, we don't have to make do with thirty year RPI rolling on a ten year basis. The thirty years is staitstically very important; I cannot see that a fixed thirty years is, particularly when the baseline starts to include distant years that are now anachronistic and far more untypical of the current norm than more recent years yet to be rolled in to the baseline. As I've said before on here; in an oscillating climate varying randomly around a more or less static long term mean a fixed historic baseline is fine, but once you get sustained change that baseline is always out of kilter. That's why chartists use time series of differing lengths; it is the interplay between different series that indicate underlying trends.

[Stratos Ferric]

I think we may get a rise of 0.2C today, however with even the BBC website citing widespread single figure minimas for the rest of the week in their monthly outlook it does look like the CET will be heading south after today. If we do hit 15.5C tomorrow it only needs to be 14.4 between then and Saturday to get back to 15.2C

Mins of around 9C and maxes of 19C to 20C are very much on the cards I would say, unless the outlook has changed since 5am this morning

I think it's those mins of which I'm dubious Stu. We never quite fall under the influence of polar air this week, however modified. Wed-Thur looks most likely for 10 or below, but that would require clear skies and still air; I can see one but not definitely both of those. I don't see a climb immediately beyond Wednesday, but assuming we're up at 15.6 by then I don't see a fall right back to 15.2 either.

[DR Hosking]

last 3 nights min in Northamptonshire 14.1, 14.5 and this morning 15.5. Cant see a big change over the next couple of days so I can see the CET rising until at least mid week, even with maxes of 21 degrees. After that looks like cooler minama and chances of falling CET.

[Stratos Ferric]

last 3 nights min in Northamptonshire 14.1, 14.5 and this morning 15.5. Cant see a big change over the next couple of days so I can see the CET rising until at least mid week, even with maxes of 21 degrees. After that looks like cooler minama and chances of falling CET.

Given that mean to date is only 15.5, it would be hard NOT to increase if the minima stayed at those levels DR. Maxima would have to stay low to avoid it. It is high overnight minima presently that are indicating a drift up to the middle of the week. Thereafter they look like falling away again to the sort of values we saw last week.

[VillagePlank]

My 16.1C looks to be in the ballpark, again. But then I said that last month about the same time ...

[West is Best]

As Mr Holmes and I are frequently at pains to point out, the important thing is to state clearly your assumptions ahead of any numerical assertions, thereafter - within reason - it's a matter of horses for courses.

Re my use of the rolling average. It is statistically a much better baseline. I would have more sympathy for UKMO if they stepped the baseline forwards immediately each ten year chunk had been completed (alowing a little time to check data, etc.). I have a suspicion that this is just an anachronism dating back to a time when calculations were more time consuming. After all, we don't have to make do with thirty year RPI rolling on a ten year basis. The thirty years is staitstically very important; I cannot see that a fixed thirty years is, particularly when the baseline starts to include distant years that are now anachronistic and far more untypical of the current norm than more recent years yet to be rolled in to the baseline. As I've said before on here; in an oscillating climate varying randomly around a more or less static long term mean a fixed historic baseline is fine, but once you get sustained change that baseline is always out of kilter. That's why chartists use time series of differing lengths; it is the interplay between different series that indicate underlying trends.

SF - unless I'm much mistaken (I may be) I was advocating precisely what UKMO are now advocating i.e. if you're going to use a quincile or tercile calibration this is done against the average rather than against an absolute. Did you check out the links I put from UKMO on this?

Of course your baseline isn't 'better'. It's 1/3rd the length. For any given month you only have 10 inputs - a very weak statistical baseline. It's so open to skewing by outliers as to be meaningless. For a given month you need a good 30 inputs - though it's actually much more complex than this as the UKMO links illustrate.

You are, if I may say so, rather disingenuously mixing this point with another one entirely - namely that they should be using the most up to date baselines. That's a different issue. I have more sympathy with that, which is why I like the idea of using the 'long term' rolling 100 year average - so that one can say 'this was the coldest in the last 100 years' etc. I've no real problem updating it every year, but it's a bit of an unnecessary fiddle. There is something rather more noble about not allowing the whims of any given month or year to blow one around. Updating the decadal average every, well, 10 years seems to me to be entirely reasonable. All the more so in this faddist age of believing every single storm, strike of lightning, iceberg or snowflake to be immediately attributable to 'Global Warming'.

Edited July 16, 2007 by West is Best

[Gavin P]

Keep in mind DR, that the CET zone is generally a little bit north and west of Northamptonshire (hence the fact there was no rise yesterday, when Norhants had a warm night on Saturday night and a humid day on Sunday)

I'll be surprised, if the CET is much higher than 15.6 or 15.7 by Wednesday/Thursday.

Edited July 16, 2007 by Gavin P

[DR Hosking]

yes but we are no more than 10 miles outside it (in W Northants) so should be fairly closely in line with it. just pointing out that the last 3 night have been very warm, if we have 15.5 degrees here I cant see it being much different in the middle of the CET zone 30 miles away.

[kold weather]

Should see another rise today with mins being generally up in the 13-15C range and maxe shave been between 21-23C in the CET region. We will probably rise upto about 15.6-15.8C by the time we reach Thursday but given we reac ha slightly cooler airmass and the fact that we switch to a WNW/NW type airflow we should start to see a very slow drop, maybe made faster if another deep depression decides to swing across the UK. I would suspect that given the outlook on the models an average month is looking unlikely, but it may be a little closer in regards to the 61-90 average.