How exceptional is this year within the UK?

[Stratos Ferric]

Instead of looking purely at the raw averages, I've taken a different cut of the CET this evening - looking at the relative ranking of months to assess this year. Before I start, let's assume that November outturns at 5.2C and December at 4.5C: this year would then come in all but level with last year; this would also be the second coldest finish to the year in the last 25 - which given the trend we're on is assuming rather too much, however...

This year would come in at no.15 by temperature, but only 23rd in terms of sum of ranks (i.e. adding together the score for each month of its rank against all its other antecedents).

What IS remarkable about this year is that it has three top 5 months, including two no.1's. Given that there are only 60 (12 months x 5) available across the whole data series this is the sort of dominance equal to the likes of Liverpool and Man United in the FA Cup and Championship. 1995, 1990 and 1976 each had two.

Only two years have landed without a single month in the lower half of the total for each month. Staistically, the odds against this are 0.5^11, i.e around 2500-1 against, so that it's happened at all is noteworthy - unless the data sample is not normal - and clearly now it is not. That these years have both occurred in the last five (2002 and 2005) is statistically very significant.

2006 is "saved" by March, which ranks 210th warmest, though this would be beaten by November on my assumption above, which would come in 252nd warmest, and in terms of rank therefore the coldest month this year.

But if we take a cut from May - October incl, the sum of ranks for 2006 is 158; on average each month is around 26th, or well inside the top 8%. This total is 36 lower than the next lowest (1959), 56 below 1947 (and remember how cold winter was that year) and 114 below 1989. The actual ranks from May have been: 57th, 27th, 1st, 69th (this is astonishing given how disappointing August semed to be in many people's minds), 1st, and, 3rd.

For consecutive warmth the year is also without peer. The sequence from July runs 1st-69th-1st-3rd.

Other years have had longer runs of top decile months (three on the bounce is not unusual and four occurred in 1933 and 1893) but none with such a concentration of extreme warmth. Summer 1976 has 5th-4th-12th (June-Aug), 1995 had 6th-1st-125th-5th starting from July; 1945 had 6th-10th-9th from Feb-Apr and 1990 had 9th-3rd-4th from Jan - the disappointing winter to top all disappointing winters!

Stepping back from the present, the last lower decile month was May 1996, coming in 334th warmest. In the 125 months since then the next coolest have been Dec 1996 (270th), Jan 1997 (236th), June 1999 (225th); only another 16 have come in as lower half months, less than 1/6th of the total, and the rate is declining. Since May 1996 there have been 39 - yes, thirty nine - upper decile months. Over 10% of all very warm months have occurred in the last ten years or so.

So, what of this year? It's hard to tell. I have already said previously that I have a hunch that this is the end of the beginning. On the one hand, in terms of actual average, this year will probably just break the overall record, but what has been astonishing is the over summer warmth. The question is, was it the summer warmth that was remarkable, or the fact that given this warmth winter managed to be comparatively cool. It's easy to see summer as the outlier, but if my hunch is correct then, I'm sorry to say, it might be winter that's the anomaly in the christmas pudding.

[AtlanticFlamethrower]

2006 is "saved" by March, which ranks 210th warmest, though this would be beaten by November on my assumption above, which would come in 252nd warmest, and in terms of rank therefore the coldest month this year.

Stratos Ferric, you are smart enough to know you can't say that. :lol:

You make a good case for this year being a measured exception on the warm side - here in UK.

However, Britain is such a small island. In parts of Russia they'd use one thermometer per Britain-size area.

Last winter was an anomaly. How to explain? Exhibit A: It's the start of a new pattern of more -ve NAO / blocked winters.

[Stratos Ferric]

Stratos Ferric, you are smart enough to know you can't say that. :lol:

You make a good case for this year being a measured exception on the warm side - here in UK.

However, Britain is such a small island. In parts of Russia they'd use one thermometer per Britain-size area.

Last winter was an anomaly. How to explain? Exhibit A: It's the start of a new pattern of more -ve NAO / blocked winters.

I can say that - the fact is that in 2006 March (ok - and Feb) have been big outliers. The next coldest month was January, which was 106th warmest - well in the top third of all time. In the context in which I used it, and with the addition of February, my use was perfectly acceptable.

Not sure what the density of thermometers has to do with anything; you'd better elaborate for me.

As for the pattern of more blocked winters, one swallow does not a summer make, nor does one cold month a long trend make. Let's not get too carried away, it'll probably only end - as it often seems to on here - in tears.

[J10]

Interesting Article, but looking at the last 6 months and average against 1971-2000 CET (From Met Office)

May 12.3c (+1.0c)

June 15.9c (+1.8c)

July 19.7c (+3.2c)

August 16.1c (-0.1c)

Sept 16.8c (+3.1c)

October 13.0c (+2.6c)

After a very wet May, there have been 4 months way above average, with the exception being August, which itself was still well above the 1961-1990 average.

If August had not been so weak in comparison with the other months, this summer/early autumn would have been even more extraordinary.

[Stratos Ferric]

Interesting Article, but looking at the last 6 months and average against 1971-2000 CET (From Met Office)

May 12.3c (+1.0c)

June 15.9c (+1.8c)

July 19.7c (+3.2c)

August 16.1c (-0.1c)

Sept 16.8c (+3.1c)

October 13.0c (+2.6c)

After a very wet May, there have been 4 months way above average, with the exception being August, which itself was still well above the 1961-1990 average.

If August had not been so weak in comparison with the other months, this summer/early autumn would have been even more extraordinary.

It would have been extraordinarily extraordinary, and it's not often you can say that, though I know for a fact there was at least one occasion when WB said it.

[AtlanticFlamethrower]

I can say that - the fact is that in 2006 March (ok - and Feb) have been big outliers. The next coldest month was January, which was 106th warmest - well in the top third of all time. In the context in which I used it, and with the addition of February, my use was perfectly acceptable.

I object to the term "outlier" - it was the weather. It actually happened. Calling it an "outlier" is suggesting it happened but only just and had itself had no impact on patterns. It is more likely that March was part of a larger pattern and did impact itself in a small way on a pattern.

Not sure what the density of thermometers has to do with anything; you'd better elaborate for me.

Statistical quirks, warm and cold. The reason why Britain's CET is derived from pool of thermometers is the principle that more thermometers = more "real" and accurate representation of UK temperature. Had there been fewer thermometers the temperature representation would have been less accurate or real. I accept the measured warming - what matters is our perception of our warming in relation to climate trends, as the point of your post was to claim last winter was an anomaly. It is probable parts of Russia have only one MET standard thermometer per Britain-sized area. It is therefore difficult to compare trends there to here both places, let alone assume they are of equal value. Wealthy, populated Britain is no more significant temperature area to climate than barren, wilderness of Siberia!

As for the pattern of more blocked winters, one swallow does not a summer make, nor does one cold month a long trend make. Let's not get too carried away, it'll probably only end - as it often seems to on here - in tears.

That's what you said and it remains to be seen if NAO will "never go -ve again, properly" which is what your argument amounts too.

[Stratos Ferric]

I object to the term "outlier" - it was the weather. It actually happened. Calling it an "outlier" is suggesting it happened but only just and had itself had no impact on patterns. It is more likely that March was part of a larger pattern and did impact itself in a small way on a pattern.

Statistical quirks, warm and cold. The reason why Britain's CET is derived from pool of thermometers is the principle that more thermometers = more "real" and accurate representation of UK temperature. Had there been fewer thermometers the temperature representation would have been less accurate or real. I accept the measured warming - what matters is our perception of our warming in relation to climate trends, as the point of your post was to claim last winter was an anomaly. It is probable parts of Russia have only one MET standard thermometer per Britain-sized area. It is therefore difficult to compare trends there to here both places, let alone assume they are of equal value. Wealthy, populated Britain is no more significant temperature area to climate than barren, wilderness of Siberia!

That's what you said and it remains to be seen if NAO will "never go -ve again, properly" which is what your argument amounts too.

AFF, in the context of this year March and Feb are unusual months, in that sense they are definitely outliers. The fact that March was the coldest month by rank since June 1999, i.e. 80 months previously, I think qualifies me additionally to use the term "outlier". Beg to differ if you like, but I'm sorry, semantically you're wrong. July also happened - are you going to tell me that wasn't an outlier either. Before responding you might want to go and check the dictionary definition of "outlier" - the fact that something happened has no relevance at all to whether or not it was an outlier, indeed, unless something DID happen it canot fall into a data set from which an outlier could be selected anyway.

I still don't understand what point you're making about thermometers and what relevance at all this has to this thread. There may be fewer thermometers in Russia, but there is lower GDP per head, higher infant mortality, more uranium, and a lot more coal. What of it? Are you somehow suggesting that CET is not representative? It doesn't mater how many sites are used; the point of the CET is that the SAME sites (more or less) have ALWAYS been used, so changes in the data are indicative of changes in the environment being measured - whatever the cause of those changes.

My point was NOT to claim that last winter was an anomaly. I did offer a point of view at then end, but it was AT THE END of the thread. Feel free to disagree with that, but I think your case would be more plausible if you put up some argument, preferably with some facts to back it up, and better still based on comprehension of what words actually mean.

And I said absolutely nothing about NAO ever going -ve again - or perhaps you'd like to pin a quote that proves I did. Nor did I say anything that could be taken even to infer such a thing.

As Churchill once said: "I do not for one moment resent criticism, even when that criticism departs from all reality".

I can only assume your distaste for the argument I did put forward relates to some life or death attachment to prospects for cold weather.

I accept the measured warming - what matters is our perception of our warming in relation to climate trends, as the point of your post was to claim last winter was an anomaly. It is probable parts of Russia have only one MET standard thermometer per Britain-sized area. It is therefore difficult to compare trends there to here both places, let alone assume they are of equal value. Wealthy, populated Britain is no more significant temperature area to climate than barren, wilderness of Siberia!

What on earth ARE you on about? What ever has SIberia, the Ukraine, or aywhere else got to do with a discussion about temperatures this year IN THE UK?

There's a village missing an idiot somewhere.

Certainly an exceptional year thus far, record warmth by some distance for the May to October period, made even more remarkable by the 'averageness' of what is often the warmest month - August.

The winter ahead could make the Summer/Autumn warmth look even more remarkable when juxtaposed against it.

...although the odds are that it won't.

[AtlanticFlamethrower]

AFF, in the context of this year March and Feb are unusual months, in that sense they are definitely outliers. The fact that March was the coldest month by rank since June 1999, i.e. 80 months previously, I think qualifies me additionally to use the term "outlier". Beg to differ if you like, but I'm sorry, semantically you're wrong. July also happened - are you going to tell me that wasn't an outlier either. Before responding you might want to go and check the dictionary definition of "outlier" - the fact that something happened has no relevance at all to whether or not it was an outlier, indeed, unless something DID happen it canot fall into a data set from which an outlier could be selected anyway.

I'm not going to refer to the dictionary to forecast the weather for me.

I object to your whole concept of an "outlier" such you can throw away a month's temperature from your analysis, which is what you implied by saying March "saved 2006."

You are saying March disguised more fundamental goings on, ie. an even higher yearly CET.

If you really want to dump March, I'll dump July and we're even.

I still don't understand what point you're making about thermometers and what relevance at all this has to this thread. There may be fewer thermometers in Russia, but there is lower GDP per head, higher infant mortality, more uranium, and a lot more coal. What of it? Are you somehow suggesting that CET is not representative? It doesn't mater how many sites are used; the point of the CET is that the SAME sites (more or less) have ALWAYS been used, so changes in the data are indicative of changes in the environment being measured - whatever the cause of those changes.

Britain is a small land area. You are using observed temperature within this land area to make staggering claims - that March was an outlier, that last winter was an anomaly, "that this is the end of the beginning". I say you've decided something is already underway and it is with reference to that thing you define what data to accept and what to chuck.

Within the land area of UK there are significant +/- anomalies - maybe as much as -/+ 1C. Multiple sources means the UKCET average is reflective of the overall temperature pattern. However - most places don't have this temperature accuracy. Whereas of the UK we may be 90% certain of temperature accuracy, France it may be 80%, Poland may be 70%, Russia 40% etc.

The point I'm making is you look at global trends through UK CET. This is stupid. You cannot bung the significance of UK's climate this year all under "the end of the beginning" - which means you believe just like UK's climate the world's climate has reached a threshold.

These hypothetical Russian thermometers could disguise significant cooling in Britain, or indeed far greater than Britain-sized areas. UK CET is fun statistic but how much we can read climate lessons into it depends on how reliable other statistics are.

My point was NOT to claim that last winter was an anomaly. I did offer a point of view at then end, but it was AT THE END of the thread. Feel free to disagree with that, but I think your case would be more plausible if you put up some argument, preferably with some facts to back it up, and better still based on comprehension of what words actually mean.

I'm not the person to do that. I objected to your use of "outlier" for something that REALLY HAPPENED and what that implied for the rest of your thinking about winter. It makes no sense to call something that happened an outlier. There is no reason a 1 in 100 years phenomenon cannot happen 2 years in a row. Irrespective of the fact we may be wrong about it being a 1 in 100 years event.

And I said absolutely nothing about NAO ever going -ve again - or perhaps you'd like to pin a quote that proves I did. Nor did I say anything that could be taken even to infer such a thing.

As Churchill once said: "I do not for one moment resent criticism, even when that criticism departs from all reality".

Cold winters will come from -ve NAO. I inferred that your "end of the beginning" meant that the beginning of the end of cold winters had past and that the end of cold winters had arrived. Therefore the end of the beginning for -ve NAO.

I can only assume your distaste for the argument I did put forward relates to some life or death attachment to prospects for cold weather.

What on earth ARE you on about? What ever has SIberia, the Ukraine, or aywhere else got to do with a discussion about temperatures this year IN THE UK?

IMO the life and death discourse seems to revolve around runnaway Global Warming theory.

As I said above, how good your judgement of weather in UK depends as much on CET data gathered in other places, which is far less accurate than UK.

Drawing conclusions from the exceptional temperatures in Summer and Autumn this year, which I agree has been very warm, is risky unless we have equally accurate knowledge of temperatures from elsewhere.

Throwing out a month such as March from for being "too cold" (ie. an outlier that "saved" UK CET from being ridiculously high) is breathtakingly arrogant way to draw hard conclusions on (future) state of the climate.

[su rui ke]

I wonder if throwing in some global data might help here. Below are the monthly worldwide (land and ocean) temperature anomalies for this year, as published on NOAA's 'State of the Climate' pages, together with each month's ranking in the series. The anomalies are relative to the 1961-90 means.

Jan +0.36C 13th warmest

Feb +0.53C 6th warmest

Mar +0.55C 7th warmest

Apr +0.45C 8th warmest

May +0.50C 5th warmest

Jun +0.60C 2nd warmest

Jul +0.56C 3rd warmest

Aug +0.54C 4th warmest

Sep +0.56C 4th warmest

I'd be very interested if anyone could explain exactly what series is being used here and whether the complete data set is available anywhere. (I don't think it's on the NOAA site.) Then we could do SF's ranking for each year based on this too!

[WhiteFox]

The point I'm making is you look at global trends through UK CET.

As I said above, how good your judgement of weather in UK depends as much on CET data gathered in other places, which is far less accurate than UK.

Without wishing to take sides here (which I realise I am) and recognising the fact that SF is more than capable of arguing his own corner (but he's probably quite sensibly asleep) I have to take you up on these points.

I cannot see anywhere in SF's posts where he is commenting on Global trends. As far as I can tell he is making a perfectly innocent point about how exceptional 2006 has been with regards to temperatures. Another perfectly innocent point is that without March being quite below average, 2006 would be far and away the most exceptional year we have seen in perhaps our lifetimes. I really don't see why this is so contentious. Surely you can't disagree that the warmth this year (although I haven't experienced it having been in the US) is exceptional?

As for the second paragraph, surely a poster of your experience cannot make the mistake of saying "CET data gathered in other places"? A desperately contrary argument if I've ever heard one, as you know full well that CET stands for "Central England Temperature" so how can it ever come from other places?

The point that I think SF was trying to make, and I'm sure he'll correct me if I'm wrong, is that compared to historical records using CET readings (the longest continuous records of temperatures anywhere in the world), 2006 has been exceptional in terms of warmth. No extrapolation was implied to the global temperature. I cannot see where the argument lies...

[Gray-Wolf]

Strangely enough I was rummaging through the MetO's CET figures for the yearly temps last night (hadn't looked that well before) and the + 10c years seemed to stand out.

Historically they turn up,either singularly or in pairs.every 7 to 14yrs until recently. Does anyone want to explain why a pattern of maybe 2 or 3 influences suddenly goes bonkers over the past 10yrs? My little lad wants an explanation and I can rough out the 'old' patterns but, apart from GW, I'm stumped to explain the last 10yrs. Surely it is too simple an answer to say GW and if this is the answer surely we should all be a lot more worried for our own lives/continuation(Stern report aside)?

[Nick F]

Strangely enough I was rummaging through the MetO's CET figures for the yearly temps last night (hadn't looked that well before) and the + 10c years seemed to stand out.

Historically they turn up,either singularly or in pairs.every 7 to 14yrs until recently. Does anyone want to explain why a pattern of maybe 2 or 3 influences suddenly goes bonkers over the past 10yrs? My little lad wants an explanation and I can rough out the 'old' patterns but, apart from GW, I'm stumped to explain the last 10yrs. Surely it is too simple an answer to say GW and if this is the answer surely we should all be a lot more worried for our own lives/continuation(Stern report aside)?

Is a slightly worrying trend, since 1994 only 1996 and 2001 have not reached the yearly 10C CET figure. Like you say, before this we had periods of upto 15 years, such as the years 1960 to 1974 where 10C wasn't reached. Never in the record of the CET has there been such a long run of consecutive 10C+ yearly CET values than recently.

The reason for the increase in above average months all-year-round in recent years maybe contributed to the shifting North of Atlantic depression tracks keeping the UK increasingly on the warmer side of the PFJ; an increase in SST temperatures in the Atlantic and seas around the UK, but this may well be an indirect cause of GW anyway, though could be due to a change in thermohaline circulations, etc - though this seems less likely a impact than GW. At the end of the day, GW seems the most plausible explanation atm.

[Stratos Ferric]

I wonder if throwing in some global data might help here. Below are the monthly worldwide (land and ocean) temperature anomalies for this year, as published on NOAA's 'State of the Climate' pages, together with each month's ranking in the series. The anomalies are relative to the 1961-90 means.

Jan +0.36C 13th warmest

Feb +0.53C 6th warmest

Mar +0.55C 7th warmest

Apr +0.45C 8th warmest

May +0.50C 5th warmest

Jun +0.60C 2nd warmest

Jul +0.56C 3rd warmest

Aug +0.54C 4th warmest

Sep +0.56C 4th warmest

I'd be very interested if anyone could explain exactly what series is being used here and whether the complete data set is available anywhere. (I don't think it's on the NOAA site.) Then we could do SF's ranking for each year based on this too!

SRK,

I'm fairly sure that the global figures don't stretch back very far - a few tens of years at most, even so there is an astonishingly consistent pattern there, and assuming that's tracked against a similar thirty year mean then it may well understate the current rate of warming.

The point that I think SF was trying to make, and I'm sure he'll correct me if I'm wrong, is that compared to historical records using CET readings (the longest continuous records of temperatures anywhere in the world), 2006 has been exceptional in terms of warmth. No extrapolation was implied to the global temperature. I cannot see where the argument lies...

WF,

Thank you for your measured assessment. Your final point is the key one. No inference should be taken, nor - clearly - was any made or inferred regarding global trends (though SRK's data is interesting). Whatever is happening with climate, the system is bigger than than the earth's landmasses, and what is happening in one place need not in any way relate to what is happening in another. At present, I am of the view that there is fundamental global change ongoing, but within that there will still, and always, be local variations.

My final ponder, and it was intended to be slightly provocative, is that we may come to look back at this year and see a point of inflexion, or at least the start of one, on the warming trend. I hope not, but suspect that we just might. I know I have said it before on here, but extraplating more rapidly upwards we would soon have winters which are little worse than the autumns of ten years ago, with little if any lowland snow, especially in the south of the UK.

Is a slightly worrying trend, since 1994 only 1996 and 2001 have not reached the yearly 10C CET figure. Like you say, before this we had periods of upto 15 years, such as the years 1960 to 1974 where 10C wasn't reached. Never in the record of the CET has there been such a long run of consecutive 10C+ yearly CET values than recently.

The reason for the increase in above average months all-year-round in recent years maybe contributed to the shifting North of Atlantic depression tracks keeping the UK increasingly on the warmer side of the PFJ; an increase in SST temperatures in the Atlantic and seas around the UK, but this may well be an indirect cause of GW anyway, though could be due to a change in thermohaline circulations, etc - though this seems less likely a impact than GW. At the end of the day, GW seems the most plausible explanation atm.

As you and G-W point out. upside "scores" are more frequent. On their own these tell only half the story though, because one hypotehsis would be that we have a more amplitudinal climate, so we need to check the other end of the scale as well.

My laptop displays about 50 lines of data, and scrolling through the screens as I type this, after applying some conditional formats to the cells with yearly CET outturn displayed in order to show shades of blue for <8.5, <9 and <9.5C, most screens are awash with blue. There's aperiod 1727-1738 with only one shaded year, and a sequence of nine without - but even in this there are plenty of cold winter months. 1933-1950 is a strange belend of very cold and not cold. 1988-2006 has but one cold year (1996). We are now in the longest sequence of years without even a slightly cold one occuring.

The inference is clear: the upward trend applies both top end and bottom end.

[BLAST FROM THE PAST]

Way too much emphasis is being put on the UK...it has very little to do with global climate.

BFTP

[Stratos Ferric]

It depends, SF, on whether or not the undisputed warming is driving the synoptics. Have we gone from a pattern of slightly warmer summers and much warmer winters to much warmer summers and only slightly warmer than average winters ?

The back end of winter 04/05 and last winter is too small a sample to draw on but for me there could have been a shift away from the strongly positive NAO winters that characterise the even larger teapot period.

You know I've been saying for 2-3 years now that we will probably have to wait until around 2010 to see where all this is headed, bbut as I say each year about now, with each passing winter that does not deliver a cold correction the outlook bcomes more bleak for winters as we remember them.

Last year I felt that autumn suggested a continuation of a pattern, albeit a slight one, that started late in 2003, with repeated potential for polar bursts behind incoming depressions. Usually a secondary feature developed, but on occasion we would get a short lived feed in the style of the late 70s / early 80s. TWS and I often remarked on this. I am seeing none of this this year at present, whcih bothers me slightly. That said, it is not the only presage of cold, and there has been an undeniable pattern in recent winters for late winter blocking; I think the UK winter actually often had late blocks, but now thatearly winter is more like a continuation of autumn, this late feature is more noticeable and much more THE defining feature of our winters.

Re your first para, I am very firmly in the camp of warming driving synoptics. How can it not; given that the macro scale system will shift, the surface must follow? One of these days I will follow up on TWS's excellent work plotting depression tracks to test the hypothesis that they have shifted north. There is a larger body of argument on here that tends to the "we're warm because the synoptics are all wrong", but we can only say that for so long. It's like standing on the beach saying "my shoes are wet because the waves are coming further in, but the tide is only rising because the waves are coming further in". Yes, yes, yes; but step back..what's the bigger picture?

Way too much emphasis is being put on the UK...it has very little to do with global climate.

BFTP

LORDY LORDY - have you actually read my first post. Okay, the thread title does not say the thread is expressly about the UK, but nor does it in any way suggest that the consideration was global. My prose is quite clearly UK only. To be clear, perhaps we should ask the mods to change the title to read - for those incapable of making the clear inference - ...in the UK.

[Thundery wintry showers]

What I think is pretty certain, though, is that if global temperatures continue rising at the current rate, if we are starting another step-change in Britain's climate for the warmer, it will have to be connected mainly with regional synoptic and SST anomaly changes (these symptomatic of a warmer globe) and not raw global temperatures.

Britain has warmed by about 0.8C relative to 1961-90, as opposed to about 0.5C for the globe as a whole, so it would require our climate to get "out of kilter" with the global average.

[Rob]

We really need to get America involved with the global warming efforts as they are the worst contributors and arent doing much to help

[Osbourne One-Nil]

...perhaps we should ask the mods to change the title to read - for those incapable of making the clear inference - ...in the UK.

You called?

[Stratos Ferric]

You called?

Your sensory capabilities are undiminished. The northern boundary of the mysterious zone of sound northern thinking is clearly secure.

[West is Best]

So, what of this year? It's hard to tell. I have already said previously that I have a hunch that this is the end of the beginning. On the one hand, in terms of actual average, this year will probably just break the overall record, but what has been astonishing is the over summer warmth. The question is, was it the summer warmth that was remarkable, or the fact that given this warmth winter managed to be comparatively cool. It's easy to see summer as the outlier, but if my hunch is correct then, I'm sorry to say, it might be winter that's the anomaly in the christmas pudding.

What an excellent post SF, which I've only just had time to read carefully. It has been an extraordinary sequence. I risk runningout of superlatives here, and don't want to slide into the political commentators' irritating habit of putting very very in front of everything (Andrew Marr and Nick Robinson: joint = worst offenders). Which is why stats like yours here speak volumes in and of themselves.

[Dawlish]

What a super debate. One of the reasons I post on here. Well done for the stat analysis SF; it would be hard to find anything else except warmth in the stats for the last 20 years - unless one thought that one average year makes a pattern change - and your novel analysis shows that very well. On the other side, well done to others for the responses.

su rui ke's contribution is so telling. Last winter was exceptional in the UK, in its own way. The coldest in 9 years. 1 in 9 merits exceptional in my book. It didn't buck any trend and it certainly hasn't started a new one. The Global Warming trend is not linear; never has been at any stage in the warming. The UK will not follow that trend slavishly (no area will) and therefore there will be even more deviations from the global trend in a given, tiny area like ours. However, the trend remains in place, as su rui ke clearly shows. The world is warming. There is no evidence whatsoever that the trend is about to reverse, only theories about how it could.

There will still be cold winters. The non-linear trend will ensure that, however 3 things will happen, if the GW trend continues: the frequency of the colder winters will reduce; the intensity of the cold will be less; the length of duration of that cold will be less. If the GW trend continues, that will be inevitable and as TWS has stated; the UK cannot be immune. Atmospheric physics would not allow it.

Paul

[The PIT]

It's been an interesting year. First of all an end of above normal months. Had to happen sometime. Then months well above normal yet split by an average month in one of the biggest reversals seen temp wise. Now all we need to cap it off is one huge storm. Anyone remember the silly warm night courtesy of the ex hurricane.

[Stratos Ferric]

It's been an interesting year. First of all an end of above normal months. Had to happen sometime. Then months well above normal yet split by an average month in one of the biggest reversals seen temp wise. Now all we need to cap it off is one huge storm. Anyone remember the silly warm night courtesy of the ex hurricane.

It might be my reliance on recall rather than data, but I don't think we get big blows like we used to. TM, TWS, OON, John, anyone else - any data? Maybe we're just in a short phase where this phenomenon is at a lower incidence, although again, another hypothesis is that the jet core is more frequently to our north, keeping the mian upward fluxes away from our shores.

D...

There will still be cold winters. The non-linear trend will ensure that, however 3 thinngs will happen, if the GW trend continues: the frequency of the colder winters will reduce; the intensity of the cold will be less; the length of duration of that cold will be less.

This is the interesting thing. I know we say it over, but your points about variation around an upwardmean do bare repeating over and over, but if there is an upward trend for the mean then unless the maxima are outperforming this trend (they don't appear to be as I illustrated previously) then the minima must be trending up as well. Your point is bang on; like it or not, winters become shorter, and within them cold spells less intense, less persistent, and less frequent on average.

Here's a reprise of a chart I produced last year...

It shows the average temperature for each deep winter month since 1960, together with a regression line for the whole series, and a 95% line, plotted using the standard deviation of the main data. I've also inserted a "by eye" minimum threshold based on actuals. The inference of that line is clear and chilling in all but consequent ambient temperature. We are ALREADY past the point when we could turn up a cold month, let alone a very cold one (other than as an occasional, and I mean very occasional - like once per 30 years, freak). Dec 95, Dec 96, Feb 96, Jan 97 - none of which were truly exceptional, already seem out of reach. Dec 81, Feb 83, Jan 85 and Feb 86 are, I am sorry to say, looking like they're not to be repeated until such time as the current warming is reversed. What is beyond question is that clustering like that is all but certainly impossible in what genuinely is starting to look like the "even larger teapot".

I know I keep saying it, but each passing winter without something startlingly cold makes the prospect of something really cold disappear further. When I was a kid I used to be transfixed by the dot disappearing on the old b&w TVs when they were turned off, getting smaller and smaller and slowly fading, until I really couldn't tell whether there was a dot there, or whether it was my imagination. Some reading this may find that this analogy is a painful one - it's no less painful for me - but it is based on facts not preference. Little did I ever imagine that the dot on the TV would become an anlogue for the disappearance of winter weather.

And before anyone throws their arms up in horror, go look at the chart above again. In RELATIVE terms (i.e. of its moment in time) last February WAS as cold as Jan 2001; it was as cold as December '95 and Feb '96; it was as cold as Jan '85 and Jan '79, and, and this is the BIG one, it was perhaps just as cold as January 1963. Save for the very very rare outliers (even these will be warmer than Jan 63 was) - the only one on the chart is Feb 86 - it is just possible that last winter is as bad as it can now get, and the trend from here is still almost certainly upwards I'm afraid.

[The PIT]

It might be my reliance on recall rather than data, but I don't think we get big blows like we used to. TM, TWS, OON, John, anyone else - any data? Maybe we're just in a short phase where this phenomenon is at a lower incidence, although again, another hypothesis is that the jet core is more frequently to our north, keeping the mian upward fluxes away from our shores.

D...

http://www.sheffieldweather.co.uk/janmon05.pdf

A few in there.

http://www.sheffieldweather.co.uk/febmon05.pdf

a few more including one unforecast one that never happened.

http://www.sheffieldweather.co.uk/feb2002wiz.htm

a few in there. So yes we still get a few blows. This year we missed them in Sheffield so far.

I wouldn't so keen to count cold months as outliers. If you do you've also got to include very warm months as outliers as well.

[Dawlish]

I wouldn't so keen to count cold months as outliers. If you do you've also got to include very warm months as outliers as well.

In a warming trend you'd have a greater number of warmer "outliers". If the trend were level, the number of colder winters would equal the number of warmer winters, over a reasonable period of time, say 20+years. The longer the time period, the closer to 50:50, warm:cold, we would get. That is patently not the case. Neither is it the case over 30, or even 40 years; the strength of the UK UK warming trend over the last 20+ years has already skewed that.

I'd rather not use the term "outlier" myself. I'd rather use the terms cold/very cold/exceptionally cold, or mild/very mild/exceptionally mild. The chances of seeing an exceptionally mild winter are increased with a warming trend. The chances of seeing a colder than average winter are reduced. The actual odds would be dependent on how far we have travelled along the trend. I've been going for 7/10 UK winters warmer than average, for a decade and a half, but the predominance of warmer winters recently is making me think that I may well change those odds to 8/10 by the end of this decade. We move further along the warming trend and the chances of a cooler than average winter decrease with each passing year. One average winter makes hardly any difference to the moving mean and thus little difference to the odds on the next winter being warmer, or cooler than average.

Paul