March CET

[nick sussex]

So at the end of the day the march CET will be below average but nothing significant which shows how sometimes the CET isnt really a true representation of the overall month.

The last 8 days really skewed the mean especially because of the high night time min however I suspect there will be wide regional variations with scotland coming out much more below average as the mild air never really impacted here. I think this month shows how difficult it is to get a very below average monthly CET whereas its much easier to get a four week period crossing over two months where the mean is well below average.

[Metomania]

So at the end of the day the march CET will be below average but nothing significant which shows how sometimes the CET isnt really a true representation of the overall month.

The last 8 days really skewed the mean especially because of the high night time min however I suspect there will be wide regional variations with scotland coming out much more below average as the mild air never really impacted here. I think this month shows how difficult it is to get a very below average monthly CET whereas its much easier to get a four week period crossing over two months where the mean is well below average.

Does your last sentence really make total sense Nick? One of the reasons I do like the CET is that when tested against a long mean it is possible to discern patterns. The problems only emerge when people read into the statistics 'meaning' based on a week, a fortnight or even an isolated month's data. So I would say the March 2006 CET is a very fair reflection of, well, March 2006.

The really interesting aspect is that March 2006 continues a trend begun last November. We have had 5 months of average to below-average months and that begins to look like a pattern.

[johnholmes]

I totally agree with your comments M. people do try to make something seem wrong when its perfectly okay. if we stop having a monthly CET, however based, and then say run it along the lines Nick seems to be suggesting, there would be no statistics at all. Certainly none that meant anything.

There are valid arguments supporting either Manley or the Met office system, or Philip Eden, but all have a monthly data set and this must be kept.

I suspect if the month had been a total reversal, ie unusually mild then a very cold spell that changed it to a below average month this complaint would not have been made.

John

[Weather-history]

I suspect if the month had been a total reversal, ie unusually mild then a very cold spell that changed it to a below average month this complaint would not have been made.

John

That is exactly what happened with November 2005, it was very mild to mid-month and then came the cold spell and it scuppered the chance of the mildest Autumn on record falling.

[nick sussex]

I totally agree with your comments M. people do try to make something seem wrong when its perfectly okay. if we stop having a monthly CET, however based, and then say run it along the lines Nick seems to be suggesting, there would be no statistics at all. Certainly none that meant anything.

There are valid arguments supporting either Manley or the Met office system, or Philip Eden, but all have a monthly data set and this must be kept.

I suspect if the month had been a total reversal, ie unusually mild then a very cold spell that changed it to a below average month this complaint would not have been made.

John

I think people have misinterpreted my thoughts here. The point I was trying to make was how difficult it is to get a very below average calendar month whereas we seem to in the last few years managed very below average four weekly periods. In no way was I suggesting that monthly CET should be ditched as this is the only way to get proper records available for easy comparison.Where did I suggest that we ditch the CET?!!!

And also last november IMO was skewed by a cold ending where the overriding memory of the month for me was mild, I'm not as some suggest obsessed by cold weather but just stating an obvious point that the CET is not always a good indicator of the overall months weather.

Edited April 1, 2006 by nick sussex

[Metomania]

just stating an obvious point that the CET is not always a good indicator of the overall months weather.

I think we had better agree to disagree :lol:

I notice looking back some arguments over which mean benchmark the current CET should be set against. Contrary to the odd opinion on here, the 1961-1990 is undoubtedly skewed towards cold. The 1960's and 1980's saw anomalously cold months and years, and this 30 year average is not in my opinion fair against the longer term CET figures. By the same token, the 1971-2000 figure is slightly above the average, skewed as it is by the anomalously warm 1990's. However, the 1971-2000 average is much closer to the long term average than the 1961-1990 figure, so I am not against using it.

However, to iron-out such arguments I do prefer a 100-year rolling average. I'd be interested in Philip's view on this, John Holmes too, as well as everyone else. It's easy to follow, and equally easy to compute. It brings the average right up to date, but avoids really anomalous cycles such as the aforementioned. The argument against is that it has never been done like that because 30-year means were considered sufficient. The counter-argument is that we are seeing some temperature extremes the like of which the planet has not experienced since records began and that this means a longer-term 100 year rolling average is appropriate. For those who are interested it usually comes in slightly below the 1971-2000 mean, but well above the 1961-1990 mean.

[Weather-history]

More than that, its going to be the coldest November-March period since Nov 95-Mar 96

Yes, it is now the coldest November-March period since Nov 95 -Mar 96 going off CET.

Depending on the values for February and March by Hadley, it could even be the second coldest November-March period since 1986-87 surpassing November 1990 -March 1991.

[nick sussex]

There are always arguments about which CET period to use and by using the 61-90 seems to be skewing things, IMO this is used by some to back theories of global warming as its obvious we will always find it more difficult to get below average months using that comparison.

At what point will the met office finally ditch that comparison? the year 2100 :lol: . Why dont they just combine both to take account of both periods and give a truer representation of recent years, it seems the use of one or the other will always cause argument.

Edited April 1, 2006 by nick sussex

[Guest Mike W]

From what I can see, it the first time since 1825 that the March CET has been 5.0 exactly. Im not pattern matching BTW, especially as their is nothing with March to match from what I can tell anyway.

[johnholmes]

hi M and Nick

The idea of a 100year rolling average, M, is not a bad idea to me. I suppose some people will say its not the way to do it. But as you commented the temperatures over the past 30-50 years have shown some unusual(for our lifetimes) differences from earlier dates, say early 20th century or even maybe the 19th century.

Hi Nick, I'm not having a go at you just trying to offer comments on how to move forward on the problem of the weather never fitting in exactly with the months. If we check back through records, and I have them for 1942'ish to 1995 for Finningley, you can see that this has always happened.

I would be interested to see a response from Philip about a 100 year rolling average.

regards

John

[Summer of 95]

And also last november IMO was skewed by a cold ending where the overriding memory of the month for me was mild, I'm not as some suggest obsessed by cold weather but just stating an obvious point that the CET is not always a good indicator of the overall months weather.

Nick has a good point here- the arbitrary start/finish points of a month are fixed by a calendar, not by weather patterns. I reckon that the following spells were substantially below average temperature wise in this part of Central England:

18 July-17 Aug 2005

20 June-19 July 2004

22 Feb-21 Mar 2006

8 May-19 June 1995 (well below that one, and more than 4 weeks long; extremely hot days in early May and late June led to both months coming out with above average CETs!)

13 Jan-12 Feb 2002 was according to Trevor Harley's site 5C above average in much of the country (I believe it regarding this area); that makes it warmer than January 1916 or February 1990: but because of the cold first week of Jan and the brief northerly around 22nd Feb neither month shows a record-breaking CET.

[Metomania]

Nick has a good point here-

I don't think it is a good point. These things iron themselves out over time. The CET is a long, very long, established record dating back to 1659 based on calendar months. Picking and choosing abitrary cold or warm spells is interesting in itself, but they cannot be set against the CET for comparison.

[Metomania]

I may have been slightly over the top about the 100 year average being closer to 1971-2000 than 1961-1990. Here are the figures:

The 3 figures are, in turn, 1961-1990 then 1971-2000 and finally the 100 year rolling 1906-2005 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

The annual CET from the 3 means is as follows:

1961-1990: 9.5

1971-2000: 9.75

1906-2005: 9.6

I cannot really see much justification for now using the 1961-1990 figures. Someone has had a familiar nip at the Met Office (!) but in fact they are moving to the more recent 30 year average. It may be true that some of those who use the 1961-1990 average do so to impress the global warming argument. On the other hand, the same might be true the other way of those using the 1971-2000 figure? I do think the 100 year rolling average lessens complaints from all quarters. It sits between the two 30 year means and that is probably a fair indication that the 1961-1990 figure is skewed to cold, and the 1971-2000 is skewed to warm. I'll start a campaign for the 100 year rollking average! Great to see John's comments, and look forward to Philip's in due course ... and others of course.

Edited April 1, 2006 by Metomania

[Summer of 95]

I don't think it is a good point. These things iron themselves out over time. The CET is a long, very long, established record dating back to 1659 based on calendar months. Picking and choosing abitrary cold or warm spells is interesting in itself, but they cannot be set against the CET for comparison.

Yes, it's a very useful record over such a long period. The point was that the raw CET figure (and the same goes for rainfall and sunshine) doesn't always give a perfect impression of the month's general characteristic. For example, just going by the CET you could imagine August 2004 being a very fine summer month with its high CET. That figure doesn't tell you however that this was due to nights being warm because of all the cloud that made it the wettest August for 40-odd years.

[Metomania]

Just to pick up on the November-March point, here are the figures against the 3 different means, assuming that February will be 3.8C and March 5C: neither published yet by the Met Office):

November 2005:

6.2C which is 0.3C below the 1961-1990 average; 0.7C below the 1971-200 average; 0.3C below the 100yr rolling

December 2005:

4.4C which is 0.2C below the 1961-1990; 0.8C below 1971-2000; 0.3C below 1906-2005

January 2006:

4.3C which is 0.5C above the 1961-1990; 0.1C above the 1971-2000; 0.4C above 1906-2005

February 2006

3.8C which is the same as the 1961-1990; 0.4C below the 1971-2000; 0.1C below the 1906-2005

March 2006:

5.0C which is 0.7C below the 1961-1990; 1.3C below the 1971-2000; 0.9C below the 1906-2005

Overall winter 2005/6:

12.5C which is:

0.3C above the 1961-1990 average (12.2C)

1C below the 1971-2000 average (13.5C)

0.2C below the 100 yr rolling 1906-2005 average (12.7C)

November 2005-March 2006:

23.7C which is:

0.7C below the 1961-1990 average (24.4C)

3C below the 1971-2000 average (26.7C)

1.5C below the 100 yr rolling 1906-2005 average (25.2C)

Since the run up to November 2005 had been a record-breaking sequence, to have 5 months below all 3 means is significant I think. The one that catches my eye is against the 100 yr rolling. It means that taking out any skewing either way, the past 5 months have been properly below average.

The question that I ask myself now is this: is this a blip in the onward march to warming, or the start of a cyclical correction?

Edited April 1, 2006 by Metomania

[noggin]

if we stop having a monthly CET, however based, and then say run it along the lines Nick seems to be suggesting, there would be no statistics at all. Certainly none that meant anything.

John

Well, I couldn't agree more. This is why I do not like statistics. They can be manipulated. I have long droned on about how observation gives a truer picture.

An example of this was in the paper the other day......a professor who had researched into the effects on humans after Chernyobl (sp?) has just said, this week, that their statistics had given a false "picture" of the effects and that they would have got a better "picture" of the effects by using observation.

I felt at one with him..... :lol:

Yes, it's a very useful record over such a long period. The point was that the raw CET figure (and the same goes for rainfall and sunshine) doesn't always give a perfect impression of the month's general characteristic. For example, just going by the CET you could imagine August 2004 being a very fine summer month with its high CET. That figure doesn't tell you however that this was due to nights being warm because of all the cloud that made it the wettest August for 40-odd years.

Exactly. Statistics do not give a true picture. That is why I do not like them. Observation is what is required.

[johnholmes]

of course we need observations. The question to ask is

at what point does an observation require to become part of a statistic.

It has to become a statistic at some point.

Even the main frmae computers from most of the model centres use MOS(Model Output Statistics).

John

[PhilipEden]

Lots of points to refer to ... I'll probably miss one or two, but hey ...

One curiosity that will interest Kevin especially, is that (based on the Manley figures, doesn't quite work with Hadley) this is, unless I've missed one, the first time since 1969-70 that every month from Nov to Mar has come in below 6°C. Another curiosity that has been talked about before, and which may now be acquiring an extraordinary symmetry, is the absence of transition periods in autumn and now spring, so that for many of us there were few if any maxes below 10°C before mid-Nov, very few maxes over 10°C between mid-Nov and last week, and the likelihood of few if any maxes below 10°C after Mar 23.

Re standard periods, I've always favoured the latest 30-year period for individual sites for all the reasons that it was established in the first place. But recognising that we are in a period of marked climate change I also, where I have space, talk about the ranking of a particular month/year/whatever over the last 100 years. So this March, for instance, was 1.4 degC below the 1971-2000 mean, and it was 29th coldest in the last 100 years. (It's similar to having a rolling-100 year average, but for the purposes of communicating with Joe Public it avoids having two different differences, which is liable to confuse).

Re calendar months sometimes giving a false impression ... yes, well, that happens all the time. I remember on 14 May 1996 Bill Giles making the crass comment that, if the second half of the month was as cold as the first half, then it would be the coldest May on record. The next day (the 15th) he said we were on course for the driest May on record, then it started raining and hardly stopped for the rest of the month. An awful lot of months are made up of contrasting periods. Remember how the sunshine last month was approaching 300% of the average after the first week? It finished on 94%.

Re having CETs for 30 or 31-day periods other than calendar months ... this is, naturally, of most interest at the extremes. Thus we have achieved a 30-day CET above 20°C between 22 June and 21 July 1976, and between 23 July and 22 August 1995, while the CET for 26 Dec 1962 and 25 Jan 1963 was close to -3°C. I've gone into these before, though I'm not sure where (!!) ... by the way, are factual threads archived on net-weather?

FWIW, there is a running 30-day CET on the "How is March Shaping Up?" page on the website and it is interesting to note that the lowest 30-day figure during winter 2005-06 was 3.2°C, compared with 3.1°C for the lowest such period during winter 2004-05.

Philip

[Weather-history]

. Thus we have achieved a 30-day CET above 20°C between 22 June and 21 July 1976, and between 23 July and 22 August 1995, while the CET for 26 Dec 1962 and 25 Jan 1963 was close to -3°C.

I saw in an old edition of Weather from the mid 1980s, an article where they looked at very cold CETs over so many days such as 5 days, 10 days etc and the coldest 30 day periods they found were

-4.4C 10/1 to 8/2/1740

-4.1C 14/1 to 12/2/1684

-3.5C 27/12/1813 to 25/1/1814

Edited April 1, 2006 by Mr_Data

[Scribbler]

Re having CETs for 30 or 31-day periods other than calendar months ... Philip

It is not only logical but sensible to keep to calendar month periods, isn't it? :lol:

I accept that extremes will often be hidden in calendar periods but if we do without that, we might as well do without the 24 hour system too.

I would suppose that the CETs could be broken down into week-long periods (for extremophiles) but to do a 30 day rolling period would not only cause confusion but would open the records to many who would say....."but if you include/remove such-and-such a date then...."

It takes several months for the Met Ofice to finalise the monthly CETs as it is!

[johnholmes]

hi

First class explanation Philip as we would expect.

I particularly liked this

'Re standard periods, I've always favoured the latest 30-year period for individual sites for all the reasons that it was established in the first place. But recognising that we are in a period of marked climate change I also, where I have space, talk about the ranking of a particular month/year/whatever over the last 100 years. So this March, for instance, was 1.4 degC below the 1971-2000 mean, and it was 29th coldest in the last 100 years. (It's similar to having a rolling-100 year average, but for the purposes of communicating with Joe Public it avoids having two different differences, which is liable to confuse).'

That seems a very good way to do things.

John

[The PIT]

Just done our figures. Average temp of 4.6C. Coldest since 1996 while the Month was a soggy with 99.7mm of rain 138% of the Average rainfall. A remarkable month overall hidden a little by the last week.

[nick sussex]

I think this CET thing could run and run as no measure seems to please everyone. So for example a scenario that could have happened last month was for arguments sake what if the last week had been even more above average and so March had come out at say average, just looking at this statistic would have suggested that the month was pretty non descript, however the truth was markedly different.

The use of the 61-90 average as a comparison to me seems to be a decision borne out of the need to prove a point about global warming and not to make a sensible comparison. The logical thing would be to use a comparison that includes as much reliable info as possible and not that cherry picked to prove a certain point.

Going back to my uni days, a statistical analysis is always better done with a bigger sample group, so the more years used weatherwise the better so why with so much data are we still using a measure that ignores the last 15 years. What is wrong with simply saying a certain month was above or below average as compared since 60 and include all the data since then.

[Metomania]

hi

First class explanation Philip as we would expect.

I particularly liked this

'Re standard periods, I've always favoured the latest 30-year period for individual sites for all the reasons that it was established in the first place. But recognising that we are in a period of marked climate change I also, where I have space, talk about the ranking of a particular month/year/whatever over the last 100 years. So this March, for instance, was 1.4 degC below the 1971-2000 mean, and it was 29th coldest in the last 100 years. (It's similar to having a rolling-100 year average, but for the purposes of communicating with Joe Public it avoids having two different differences, which is liable to confuse).'

That seems a very good way to do things.

John

I agree completely. Apologies for repeating Philip's original as well, but it bears repeating. Using the latest 30 year rolling + showing a comparison to the last 100 years seems an excellent way of doing it without causing confusion. For that reason alone it is probably high time the 1961-1990 figure was put to bed.

[Thundersquall]

When was the last time the March CET was at or below this year's value (5.03°C)?

Edited April 1, 2006 by Thundersquall