« Previous | Main | Next »

Regional and global climate update

Paul Hudson | 17:10 UK time, Thursday, 4 October 2012

Provisional Met Office statistics show that September across the UK was wetter than normal with 112mm, or 117% of the long term average.

It was also cooler but sunnier than normal.

Regionally, the North of England was much wetter than the rest of the country, with 133mm of rain, which is 165% of average, making it the wettest since September 2000.

It was also the 6th successive month where rainfall was above normal.

There were contrasts across our region though.

Yorkshire received 125mm or 170% of average rainfall, but Lincolnshire was much drier, with only 43.4mm or 80% of the long term average.

The unsettled theme looks set to continue, with more rain tonight and early next week, which could be potentially heavy, although the weekend is looking fine and dry.

Normally at the start of the month I update the UAH satellite global temperature for the previous month, for no other reason than it's published before any of the other measures.

But for technical reasons, which you can read about here, this is not yet available.

Follow me on twitter @Hudsonweather

Comments

  • Comment number 1.

    When will the BBC be reporting on autumn arriving earlier and earlier then? *silence*

  • Comment number 2.

    Piers Corbyn interviewed by Alan Tichmarsh!

    http://player.stv.tv/programmes/alan-titchmarsh/2012-10-03-1500/

    17.30 onwards.

    Being non-PC I could not possibly comment!

  • Comment number 3.

    I see Al Gore is putting his money where hiw mouth is

    http://www.thestreet.com/story/11727215/1/al-gore-walks-away-from-green-energy.html

    "Al Gore Walks Away From Green Energy"

  • Comment number 4.

    "... for technical reasons..."

    Funny that these technical faults were only discovered during a period of warming.

    Is Spencer now saying that we can scrap the whole UAH database as a reliable global temperature record; or should we just chuck the warming parts?

    Highly peculiar.

    The truth will out.

  • Comment number 5.

    "UAH Global Temperature Update for September, 2012: +?.?? deg. C"

    October 4th, 2012 by Roy W. Spencer, Ph. D.

    "I’ve been receiving an increasing number of e-mails asking, basically, is there something wrong with the Aqua satellite daily global temperatures which are posted at the NASA Discover website?

    Well, John Christy and I are ready to say, “yes, there is”. "

    Read it all:-

    http://www.drroyspencer.com/2012/10/uah-global-temperature-update-for-september-2012-deg-c/

  • Comment number 6.

    Does anyone understand what is going on with UAH?
    I quote from the site:
    "The results suggest that there has been a spurious warming in Aqua AMSU LT which has reached close to 0.2 deg. C last month. It has been increasing over the last couple years. Do NOT expect the long term warming trend during 1979-2012 to decrease, though, because there are other changes to the long-term time series which cancels out the recent spurious warming."
    Does this mean that there is warming in the AQUA CH5 figures which are *not* reflected in the UAH anomalies?
    Otherwise, how can removal of the "spurious warming" not affect the long term trend?
    Is he only going to correct this in the September anomaly?
    Confused - you wll be!

  • Comment number 7.

    Regarding the U.K. figures for September, while the overall UK rainfall figure was nothing exceptional, at 117% of "normal", the HadUKP daily rainfall figures for England & Wales are interesting.
    These show a daily rainfall figure of 27.19mm for the 23rd, 18.79mm for the 24th, and 10.72mm for the 25th, a total of 56.7mm over the 3 days, which compares with the figure of 25.04mm for Sept. 19th 1981 and 38.25mm for the period Sept. 17th to 19th 1981, the period of the last similar period of low pressure which was quoted recently.
    So while I haven't looked at how this fits in with the long-term pattern, in terms of rainfall this storm does seem to have been more intense than the one in 1981.
    Nothing significant in that of course!

  • Comment number 8.

    Recent storm data has shown that hurricane/typhoon/tornado frequency has fallen, despite the media reports to the contrary. Nothing sells like bad news.

  • Comment number 9.

    @6 QV

    "Does anyone understand what is going on with UAH?"

    I don't and after reading the latest post not sure they do. Will just have to sit back and wait, should spark some very interesting discussions across the blogosphere.

    I have already learnt about a few new bits:-

    "Calibration"

    "The MSU and AMSU instruments were intended for day-to-day use in weather forecasting and thus were not calibrated to the precision needed for climate studies......"

    http://www.ssmi.com/msu/msu_introduction.html

    and more

    http://www.ssmi.com/msu/msu_data_validation.html

    Will now sit back and await developments

  • Comment number 10.

    6. QuaesoVeritas wrote:

    "Does anyone understand what is going on with UAH?"

    Apparently not, including Roy Spencer.

    It looks like he's not going to update the UAH database this month, but rather wait until their version 6 is ready in a few weeks then re-issue the whole set.

    Possibly the UAH value for September, as it stands, does something nasty to his nice polynomial trend line, that he adds 'for entertainment purposes only'?

    Like you, I don't see how, if the current noise is purely a warming signal as Spencer appears to say, it won't affect the trend. If it were an 'amplification' signal, perhaps caused by water vapour fluctuations in the LT due to ENSO, then that would make sense, because cooling caused during La Nina would also be exaggerated.

    Satellite data has long been suspected of being overly influenced by ENSO in any case.

  • Comment number 11.

    A UAH anomaly figure of 0.34c (0.338c) for September, the same as the original one for August, but that has been reduced to 0.21c (0.208c), along with most of the figures since 2010.
    As far as I can tell, the long-term trend has changed from 0.136c/decade to 0.131c/decade at the end of August.

  • Comment number 12.

    11.QuaesoVeritas:

    Very peculiar goings on.

    In any case, a very big jump between August and September, and September 2012 is joint 3rd warmest in UAH record. You have to go back to July 2011 to find an equally high monthly anoamally, and back to September 2010 to find one that was higher.

    Interestingly the 30 year trend reduces slightly but remains at +0.16C to 2 d.p. per decade; the same as every other global data set we have (+/-0.01C). The new revisions lower the UAH monthly anomalies by around -0.05C on average since 2010.

    Roy Spencer will be pleased to note that his beloved polynomial trend line (which has no predictive value) is downwards again. That should please the readership of WUWT, for whom I suspect it is chiefly intended. Try adding a 5th order polynomial to it instead of a 4th order one.

  • Comment number 13.

    #12. - newdwr54 wrote:
    "Interestingly the 30 year trend reduces slightly but remains at +0.16C to 2 d.p. per decade;"
    And to a single decimal place, they are still 0.2c/decade, but to zero decimal places, they are zero! ;-)

  • Comment number 14.

    I think Spencer has made it clear that he only adds the polynomial trend to wind people up. Those that are predisposed to be wound up that is. It seems to achieve this goal admirably.
    Shame that he's having a bit of bother with that NASA satellite. They don't make 'em like they used to - just can't get the staff.

  • Comment number 15.

    #14. - lateintheday wrote:
    "I think Spencer has made it clear that he only adds the polynomial trend to wind people up."
    When did he make that clear?
    I am not sure that he has.
    He used to say it was "for entertainment purposes only", but I am not so sure.

  • Comment number 16.

    QV, I can't say exactly, because he's constantly challenged (or admonished) for adding the polynomial trend, but my gut tells me it was early this year. Would have been at his own or the WUWT site. Seem to remember him conceding that it entertained him, in a rather childish way (by his own account) knowing that it annoyed some people.
    My memory isn't what it used to be . . sigh.

  • Comment number 17.

    Pingosan #1

    Autumn arriving earlier? Depends what you mean by autumn.

    By the reckoning of weather and temps. so far it could be said to be early - with a cool September and with October, as yet, showing no sign of being more than average at best. All following on from a dull cool summer.

    Ironically, this might give the impression that autumn is late. Fruit and seeds slow to ripen. Autumn flowers behind schedule. Lingering insects trying to catch up on delayed life cycles - which if they are all lucky will not be curtailed by an early cold spell and frost.

    Given temp levels at the moment, frost does not seem out of the question and there seems little sign that the heat missing for much of the growing season will be made up now. In any case, as late as October, only the most phenomenal warmth will come near to achieving this.

  • Comment number 18.

    Greensand#2

    Well, what can you expect from Tichmarsh? Talk about scraping the bottom of the controversy barrel to raise interest in his programme. But then he's so "lovely and nice" isn't he?

    I wonder how much they paid Corbyn? Or was it the other way round do you think!

    Incidentally - was the feller on the right an astrolloger or an astronomer? He was labelled as both at different points! Probably makes no difference. I always thought it took millions if not hundreds of millions of years for the "spiral arms" of the galaxy to revolve let alone passing in and out of cosmic dust clouds therein in periods of decades!

  • Comment number 19.

    @18 jk

    "Well, what can you expect from Tichmarsh?"

    Nowt!

    "I wonder how much they paid Corbyn?"

    Nowt?

  • Comment number 20.

    14. lateintheday wrote:

    "I think Spencer has made it clear that he only adds the polynomial trend to wind people up."

    I think he does it for the WUWT audience. His monthly UAH update is always carried by WUWT. It has been suggested that Spencer first took up adding polynomial trends to his popular version of the UAH data after correspondence with Canadian oil industry engineer Allan MacRae back in autumn 2008: http://deepclimate.org/2009/04/09/the-alberta-oil-boys-network-spins-global-warming-into-cooling/

    MacRae had helped a WUWT guest poster to use a high order polynomial trend on a temperature graphic, which, back then, gave the UAH data the following look: http://icecap.us/images/uploads/uah7908.JPG

    MacRae used that graph, which is a 6th order polynomial, to make the ludicrous claim that "Since just January 2007, the world has cooled so much that ALL the global warming over the past three decades has disappeared!": http://icecap.us/index.php/go/joes-blog/is_this_the_beginning_of_global_cooling/

    Shortly after that 'ICECAP' post from MacRae went up, Spencer started adding a fourth order polynomial trend to the UAH data published on his site and at WUWT. Fourth order polynomial is less grossly distorting of the data, but it still gave the recent UAH data a nice downward look at the time. (I don't remember ever having seen a chart of UAH data with a linear trend featured at WUWT, even though it's the linear trend that UAH publish monthly with their official data update.)

    I'm sure QV, greensand, and possibly yourself can quite easily graph the latest UAH data on excel and add a sixth order polynomial trend to it. If not, I'll happily talk you through the process. Let's just say that I seriously doubt that Allan MacRae is still referring to a sixth order as "the best polynomial" fit for the UAH data.

  • Comment number 21.

    #20. - newdwr54 wrote:
    "It has been suggested that Spencer first took up adding polynomial trends to his popular version of the UAH data after correspondence with Canadian oil industry engineer Allan MacRae back in autumn 2008"
    I thought it was only "sceptics" who believed in conspiracy theories?

    "I'm sure QV, greensand, and possibly yourself can quite easily graph the latest UAH data on excel and add a sixth order polynomial trend to it."

    I'm not sure why I would want to do that?
    All trends, whether linear or polynomial are "strictly correct", but don't really prove anything. There is nothing "magical" about them and by definition they represent the past and not the future. If we were to believe the 6th order polynomial, we are about to enter another ice age.
    I wonder if MacRae only went up to a 6th order polynomial because that is as far as you can go on Excel?

  • Comment number 22.

    21. QuaesoVeritas wrote:

    "I thought it was only "sceptics" who believed in conspiracy theories?"

    LOL. But take a look at the evidence posted in the first link.

    - January 2008: in a guest comment on WUWT, Spencer acknowledges correspondence with MacRae: http://wattsupwiththat.com/2008/01/25/double-whammy-friday-roy-spencer-on-how-oceans-are-driving-co2/

    - September 2008: MacRae posts his nonsensical 'ICECAP' claims (third link above).

    - October 2008: Lorne Gunter uses the sixth order polynomial graph from the MacRae 'ICECAP' piece (second link above) to help him claim that "30 years of global temperatures go poof". This was, apparently, also posted at WUWT, though I can't find a link.

    - November 2008: a polynomial trend line showing a latterly downward trend appears on Spencer's UAH graph for the first time.

    But yes, that might all just be coincidence.

    While it's true that no trend line is a reliable indicator of future direction, a 30 year linear least squares fit will react a lot more slowly to recent short term variability than will a high order polynomial. Polynomial trends act to smooth the longer term data and attach undue weight to more recent variations. The higher up the scale you go, the more exaggerated this distortion becomes.

    Go ahead and add a sixth order polynomial to the most recent UAH update. I promise you, it no longer looks like we're headed for the next ice age (though we might be in actuality, because the line is meaningless anyway, just as it was when MacRae was using it to make his spurious, yet confident, case for imminent global cooling).

  • Comment number 23.

    'When we are forecasting the weather, we can never be certain about:

    The current state of the atmosphere

    We only have a limited set of observations, each of which has instrumental errors and is not necessarily representative of the larger-scale circulation

    Atmospheric processes

    Our atmospheric models are necessarily only a crude simulation of what is happening in the atmosphere

    So, there is no hope of being able to produce detailed and precise forecasts

    We can make a “best estimate”, but we always need to bear in mind the uncertainties in our forecasts.'

    Dr Richard Swinbank, Met Office:
    http://icdm2012.csp.escience.cn/dct/attach/Y2xiOmNsYjpwZGY6MjczNzM=

  • Comment number 24.

    Ask yourself why the MO needs to update the shipping forecast 4 times a day.

  • Comment number 25.

    #2. - greensand wrote:
    "Piers Corbyn interviewed by Alan Tichmarsh!"

    I just got around to watching this.
    Mark Thompson was described as "a stargazer and weather watcher" by Titchmarsh, and as an astrologer and astronomer by the captions. It appears from his website that he is a genuine astronomer, but makes claims to be able to forecast weather by astronomy, about which I am sceptical although I haven't looked at his work in detail.
    http://www.markthompsonastronomy.com/
    A lot of people in the media don't know the difference between astronomy and astrology, which isn't surprising, since even some astrologers often conflate the two, in order to give their work greater credibility. It doesn't surprise me that at the level of those who put up the captions, there is confusion.
    Of course, Corbyn was able to get away with the claim that he had an "85% success rate", which is a completely meaningless figure unless you know how it was calculated. I wonder where that figure came from, Corbyn himself?
    He claimed to be able to predict events on the sun which influence the weather, but there have been instances where "unexpected events" on the sun have allegedly caused his forecast to be wrong.
    Titchmarsh himself seems to be a slight sceptic on "climate change", so it isn't surprising that he should take the line he did in his programme.
    http://www.telegraph.co.uk/comment/7111356/Alan-Titchmarsh-Climate-scientists-should-stick-to-the-facts-and-not-use-guesswork.html

  • Comment number 26.

    23. ukpahonta:

    All those statements are true in respect to whether; but not necessarily for climate.

    It's the old saying 'Climate is what you expect; whether is what you get'.

  • Comment number 27.

    An interesting edition of "Farming Today" from the BBC, about floods and land use etc.:

    http://www.bbc.co.uk/iplayer/episode/b01n5z2q/Farming_Today_Farming_Today_This_Week/

    Note, the first 2m 30s is a different programme, "Ramblings".
    It appears that in addition to problems caused by building on agricultural land, what is left also has problems in that it contains less organic material than in the past, which means that the water is not absorbed as quickly as it was. Also, there are a lot less flood plain meadows than in the past.
    Note that at the start of the programme, when Professor Simmons is asked whether it is true that we are having more "rainfall events" and more flooding, he completely fails to answer the question, instead he says what has been predicted.

  • Comment number 28.

    #26 newdwr54

    Whether its weather or climate it's the same science behind the models, isn't it?

  • Comment number 29.

    28. ukpahonta:

    I don't know enough about modelling to answer that ukpahonta.

    But weather is a very specific sub-set of climate. It is short term, highly variable and rather chaotic.

    I don't think it's a good argument to label long-term climate forecasts unreliable just because short-term weather forecasts are often wrong. (This was underlined for me this weekend; I'm painting my house. Friday was *not* 'mostly dry after mid day' where I live, I can assure you. And I haven't ruled out the possibility of legal action against the BBC/Met Office!)

    Forecasting weather is like forecasting the result of a team's performance in a single match; forecasting climate is like predicting the team's likely performance over the season as a whole.

  • Comment number 30.

    Greensand~19 /QV

    Well, quite. Only Alan Tichmarsh could make one of the greatest controversies of our age sound about as important as deciding whether to shop at Sainsbury's or Tesco.

    Do you suppose in their wildest dreams, the producers are hoping the Pres. of the Royal Society or some other will storm in to demand a right of reply - thus boosting the credibility and ratings of lovely Alan's hard hitting and ground breaking show?

    Might upset the mainly post menopausal fan base though. Better stick to Geraniums.

  • Comment number 31.

    #29 newdwr54

    'Forecasting weather is like forecasting the result of a team's performance in a single match; forecasting climate is like predicting the team's likely performance over the season as a whole.'

    For the same model the more games that you get right the better the chance of prediction for the teams final position in the league, that resolves.

    The prediction of the league position should be tuned by the accuracy of the game prediction, that resolves.

    The prediction of the league position does not determine the outcome of individual games, that resolves.

    I rather like your analogy, not so sure that others will though.

  • Comment number 32.

    New research by the University of Reading blames our recent "wet summers" on the Atlantic Multidecadal Oscillation.
    http://www.bbc.co.uk/news/science-environment-19848112
    I read this as an attempt to explain why some recent summers haven't been dry, as predicted by "climate change" theory.
    But isn't it an admission that the warming of the N. Atlantic SST and presumably atmospheric temperatures is due to natural cycles and NOT to "climate change" caused by greenhouse gasses?
    There was an interview with Prof. Sutton on the BBC R4 Today Programme this morning at about 07:25
    http://www.bbc.co.uk/iplayer/episode/b01n61s1/Today_08_10_2012/
    Actually accordign to HadUKP, which goes back to 1766, the long-term trend in England & Wales summer rainfall is down, but there have been periods of higher trends, all followed by rapid declines.
    The highest positive trend over a 10 year period was +15.3 mm/year in 1877, followed by a figure of -24.2mm/year in 1887. By comparison, the current 10 year trend is +11.77mm/year, although it's not possible to say if it has peaked.
    These swings may well correlate with fluctuations in the AMO, but I am not sure if suitable data are available for the latter.

  • Comment number 33.

    32. QuaesoVeritas wrote:

    "...isn't it an admission that the warming of the N. Atlantic SST and presumably atmospheric temperatures is due to natural cycles and NOT to "climate change" caused by greenhouse gasses?"

    One doesn't necessarily preclude the other.

  • Comment number 34.

    Well, I have obtained monthy data for the N. Atlantic from the following website:
    http://www.esrl.noaa.gov/psd/data/timeseries/AMO/
    I then compared the average summer N. Atlantic temp. with the corresponding total England & Wales summer rainfall figures, and I can find very little correlation between the two. A scattergraph using the summer N.A. temp. v rainfall figure, produced a relationship of y = 0.88x +202.9, with an R2 of only 0.0000126, where x = the absolute temp.
    Otherwise, an increase of 1c in the average temp. equals an increase of only 0.88mm of rain. Bear in mind that the N.A. temp. varies between a min. of 21.69c in 1913 and a max. of 22.92c in 2010.
    I tried using the average AMO index, instead of the absolute temp. and got a formula of 13.384x +222.12, with an R2 of 0.0023, where x is the AMO index, a slightly higher correlation, still very low. Otherwise an increase in the index of 1, results in an increase in rainfall of 13.384, bearing in mind that the index ranges from -0.478 to +0.556.
    I think it is also noteworthy that a N.A. temp. of 21.899c produced the highest summer rainfall of 409.7mm, in 1912, while one of 22.689c produced the lowest of 66.9mm in 1995.
    If anyone can see any flaws in the above, please let me know.
    One possible criticism might be that I have used the summer N.A. temp and the summer rainfall, with no offset, because I would have thought that the effect of warm sea would be more or less immediate.

  • Comment number 35.

    Wow! Just looked at AMSU-A (Ch5) temperature anomoly for October
    A significant departure from previous trends compared back to 2003
    Have I missed your comments on these trends QV?

  • Comment number 36.

    #35. - NeilHamp wrote:
    "Wow! Just looked at AMSU-A (Ch5) temperature anomoly for October
    A significant departure from previous trends compared back to 2003
    Have I missed your comments on these trends QV?"

    Have you seen Roys Spencer's recent comments about "spurious warming" in ch5 and adjustments to past UAH figures?

    http://www.drroyspencer.com/2012/10/uah-v5-5-global-temp-update-for-sept-2012-0-34-deg-c/

    On Oct.1st I posted on a previous topic (#62) that the UAH figure for Sept. looked like being between 0.4c and 0.48c, an increase of 0.06c to 0.14c compared with August, based on AQUA CH5 at the time.
    http://www.bbc.co.uk/blogs/paulhudson/2012/09/all-eyes-on-york-after-worst-s.shtml
    Of course, since then, Roy Spencer has reduced his figures for most of the last two years, including a reduction of 0.13c in the August 2012 figure and has posted a figure of 0.34c for September. If you add 0.13c to 0.34c, you get 0.47c, which is within my estimated range.
    What still puzzles me is that, my recent estimates of UAH based on previous AQUA CH5/UAH figures was coming out too low, and I was having to "adjust" my estimate by up to 0.1c. This is cannot be explained by the "spurious warming" in AQUA CH5, because I was using the latter to estimate UAH.

    I have just noticed that RSS, which doesn't apparently use AQUA CH5, has just reported a global anomaly figure of 0.383c for September, an increase of 0.128c, compared with August. This compared to my estimate of 0.45c for Sept. RSS, based on past CH5/RSS figures, which does seem to support the CH4 "spurious warming" claim. The actual increase in RSS is similar to my estimated increase in UAH.

    It is true that currently, AQUA CH5 is running about 0.533c above average, but I don't know how reliable that is in the light of Roy Spencer's comments. It does seem to have turned down over the last few days.

  • Comment number 37.

    UAH numbers for Sep are here. Note they are now version 5.5.

    http://vortex.nsstc.uah.edu/data/msu/t2lt/uahncdc.lt

    Incidentally, I am now running a page for all four global temp datasets (GISS,HADCRUT,UAH and RSS), which will update each month.

    http://notalotofpeopleknowthat.wordpress.com/2012/10/08/hadcrut-update-august-2012/

  • Comment number 38.

    This new, well it is to me! Metcheck are ramping up the prospect of storms in the not too distant future:-

    "Atlantic Set To Roar!"

    "If you ever need any proof that the Indian Summer won't happen then take a look at the jet stream and pressure chart forecasts below..."

    "....Although no damaging storms are yet on the radar, it's worth remembering what happened across the UK 25 years ago next week...."

    "...Winter is just around the corner... and the Atlantic is about to give us a heads up!"

    http://www.metcheck.com/UK/

    I have always liked the way Metcheck produce their local daily outlooks, like all only good for a couple of days, just their numerical approach/layout suited.

    But the long term announcements seem new? Or have I not been paying attention? Anyhow interesting to watch and time will tell if there is any merit.

  • Comment number 39.

    QV - regarding Spencer's adjustments. Your figures back up what Spencer said, namely that there wouldn't be much change in the long term trend. Have you by any chance looked at how the changes effect the shorter, rolling trends that you sometimes calculate?
    I was wondering whether these recent adjustments really are more in line with the other data sets and whether by using this comparison, we could have either more or less confidence in the changes themselves.

  • Comment number 40.

    39.lateintheday:

    I know your question was to QV, but I hope you don't mind if I butt in.

    UAH have now published their updated data set (V5.5): http://vortex.nsstc.uah.edu/data/msu/t2lt/uahncdc.lt

    The long term trend (since Dec 1978) is still +0.14C per decade. When Spencer posted the revised data on his monthly blog post (the one that is carried by WUWT), he only updated it back to 2010. This had the effect of reducing the long term trend to +0.13C/decade, as QV pointed out. But the complete 5.5 set confirms that it is still +0.14C; as it was before the revision (Spencer said that would be the case).

    I can confirm that these revisions are much better aligned to the other data sets for recent temperatures. This is especially true for RSS; UAH's rival satellite temperature data provider. Looks like Ch.5 is off on a tangent; even so, the warming data recorded so far in October is likely to be reflected in real world temperatures, though perhaps not to the extent suggested.

    I have an interesting hour's work ahead, copying and pasting the new UAH data onto my annual/monthly data set. I wonder whether the 198/20190 gap will widen or close?

  • Comment number 41.

    #39. - lateintheday wrote:
    "Your figures back up what Spencer said, namely that there wouldn't be much change in the long term trend."
    Which figures?

    "Have you by any chance looked at how the changes effect the shorter, rolling trends that you sometimes calculate? "
    No, I was waiting to see how/if the longer term figures have been changed.
    Now that these have been published, I will have a look, but everything seems to be happening at once. The MO have now published HadCRUT4 figures to August 2012.

  • Comment number 42.

    #40. - newdwr54 wrote:
    " But the complete 5.5 set confirms that it is still +0.14C; as it was before the revision (Spencer said that would be the case)."
    As I have pointed out before, it would take a huge change in short-term temperatures to change the long-term trend at the 2nd decimal place.

    "I have an interesting hour's work ahead, copying and pasting the new UAH data onto my annual/monthly data set."
    Same here!
    Added to which, HadCRUT4 has now been extended to August 2012!

  • Comment number 43.

    Another way of looking at the relationship between the AMO and England & Wales rainfall.
    I sorted the N. Atlantic temperatures from lowest to highest and compared them with the summer rainfall figures. I think that you would expect to see an increasing trend in the rainfall figures, but it is virtually non-existent at y= 0.0047x + 222.15, where x = the NA temp.
    In this sequence, the highest E&W summer rainfall figure of 409.7 mm, corresponds to the 15th highest temp. out of 157 and the second highest rainfall of 409.2 mm corresponds to 65th highest temp.
    On the other hand, the lowest summer rainfall figure of 66.9 mm, corresponds to the 150th warmest AMO out of 157.
    I think that you would also expect the 10 year MA summer rainfall to show a gradual increase over the temperature data when it is sorted from lowest to highest, but in fact summer rainfall is at it's highest at point 18 of the sequence and by point 157 it is in the lower half of it's range. The 30 year average summer rainfall peaks at point more or less in the middle of the temperature range.

  • Comment number 44.

    42. QuaesoVeritas wrote:

    "Added to which, HadCRUT4 has now been extended to August 2012!"

    It's too much too soon QV.

    I only just got through painting the outside of the house. (I've decided not to proceed against the BBC/Met office; apart from Friday their weather report was spot on.)

  • Comment number 45.

    40. I wrote:

    "I wonder whether the 198/20190 gap will widen or close?"

    Sorry, that should be 'the 1998/2010 gap'. Somehow a loose '9' migrated on me.

    I'm interested to see what changes, if any, this revision causes to the existing records. UAH have 1998 as the warmest year; while all the most recent surface temperature data sets have 2010 as the warmest year.

  • Comment number 46.

    #40. - newdwr54 wrote:
    "The long term trend (since Dec 1978) is still +0.14C per decade."

    According to my calculations, the trend over the entire series to August, has gone from +0.136278c/decade to +0.137268c/decade.
    I find this surprising and it appears to be due to the fact that the majority of figures prior to Sep. 2004 were reduced by about 0.01c, those from Oct. 2004 to May 2010 have mostly been increased, by up to 0.06c and those after that have been mostly reduced.
    Now can someone explain why the fact that aqua ch5 has had "spurious warming", has resulted in an INCREASE in the figures between 2004 to 2010?
    Clearly CH5 must have been showing "spurious cooling" during that period!

  • Comment number 47.

    #45. - newdwr54 wrote:
    "Sorry, that should be 'the 1998/2010 gap'. Somehow a loose '9' migrated on me."
    I was building up the courage to ask what you meant!

    "UAH have 1998 as the warmest year; while all the most recent surface temperature data sets have 2010 as the warmest year."
    That still seems to be the case.
    HadCRUT3 has 1998 at 0.517c and 2010 at 0.499c, but HadCRUT4 has 1998 at 0.52c and 2010 at 0.53c. Then one of the reasons for changing to HadCRUT4 was probably to fix that little anomaly. Come to think of it, that was probably why they stopped at 2010! It seems that the HadCRUT4 annual data file hasn't yet been extended to 2012.
    Of course that also retrospectively helps with their prediction that "about half" of the years between 2010 and 2019 would be warmer than 1998.

  • Comment number 48.

    Does the new 3dp HadCRUT4 produce a small increase in the 1850 to 2010 trend over the original 2dp version? Also seems to be a varience pre and post 1940:-)

  • Comment number 49.

    47. QuaesoVeritas:

    You have a very suspicious mindset QV. It's a good thing.

    So the UAH trend since December 1978 has actually 'risen' slightly after the latest revisions. Hmm? Also, Spencer did warn us that 5.5 was just an interim database.

    Satellite LT data are inferred from microwave soundings, and as such they rely on accurate instrument calibration and known rates of decay.

    The IPCC uses HadCRUT ground based temperature station data. As you point out, HadCRUT has undergone a fairly major update too in the past year or so. I can't remember when the previous update to the HadCRUT data was?

    UAH has now issued 5.5 version: that's 55 versions since December 1979. Another is due in the next few weeks or months.

    If anything, then at least this major UAH revision reminds us that satellite data are not the pristine, unquestionable 'ideal' that Joe Bastardi and co often claim they are.

  • Comment number 50.

    49. newdwr54 wrote:

    "I can't remember when the previous update to the HadCRUT data was?"

    Maybe, but it can be claimed that it is updated every month, depending on the station data that is available. There are also wider ranging updates:-

    http://www.metoffice.gov.uk/hadobs/crutem3/jan_2010_update.html

  • Comment number 51.

    #49. - newdwr54 wrote:
    "UAH has now issued 5.5 version: that's 55 versions since December 1979. Another is due in the next few weeks or months."

    I'm not sure if that is correct.
    Wasn't the last version v5 and the next planned one is v6, so they seem to go up in integers and v5.5 is an "interim" version.
    On that basis, there may have only been 6 versions so far, possibly more, if there were other interim versions, but I couldn't find a history of the versions anywhere and I don't really have time to dig.
    For there to have been 55 versions, that implies 9 interim versions for every version, which seems unlikely.

  • Comment number 52.

    thanks to both newdwr54 & QV for their input/comments on Spencer's revised UAH set.

    Greensand's link made me wonder if the error bars on datasets are automatically revised at the same time - presumably so. If that's the case, has anyone ever looked at whether, say, version 4 of any given dataset is still within the error bars that were originally given in say, version 1?
    Just wondering how much some of these sets have changed over time.

  • Comment number 53.

    @52 lateintheday wrote:

    "Just wondering how much some of these sets have changed over time."

    Along that line have a look at:-

    http://www.metoffice.gov.uk/research/climate/seasonal-to-decadal/long-range/glob-aver-annual-temp-fc

    "Figure 3: The difference in coverage of land surface temperature data between 1990-1999 and 2005-2010. Blue squares are common coverage. Orange squares are areas where we had data in the 90s but don't have now and the few pale green areas are those where we have data now, but didn't in the 90s. The largest difference is over Canada."

    I asked "Why do we no longer have the Canadian land surface temperature data?" Never got an answer, someone from Canada stated that there had been no changes in station locations or availability, but hey ho, there will be a reason.

  • Comment number 54.

    The HadCRUT4 annual file has now been updated and the figure for 2010 has now been increased to 0.54c, and 1998 increased to 0.523c, thus increasing the difference between the two years.
    According to HadCRUT4, 2012 is currently running at 0.419c, whereas HadCRUT3 has it at 0.379c.
    If HadCRUT3 increases by the average of UAH and RSS, i.e. 0.129c, that will make September HadCRUT3 0.637c and the 2012 average 0.416c, while the Sept. HadCRUT4 would be 0.655c and the 2012 average 0.446c, although not using the same method of calculation as the MO. However, I am not sure how valid it is to apply the UAH/RSS increase in September to HadCRUT. Based entirely on the "spuriously warming" AQUA CH5, HadCRUT3 will only be about 0.45c, and HadCRUT4 about 0.47c, representing falls on August, which leaves me very confused.
    I wonder how long the MO will continue to publish HadCRUT3.

  • Comment number 55.

    48 Greensand:What does Hadcrut4 show now that it's been updated?

  • Comment number 56.

    @55 Boanta

    "What does Hadcrut4 show now that it's been updated?"

    Not had chance to have a good look, but I am sure that QV @ 54 above will have it nailed down. Also see:-

    "An Updated Hadcrut4 – And Some Surprises"

    http://www.thegwpf.org/an-updated-hadcrut4-and-some-surprises/

  • Comment number 57.

    51. QuaesoVeritas:

    I based the '55 versions' notion on the fact that the previous UAH version was v5.4 and the latest is v5.5, so it seemed to me that they just go up in incremental steps? Apologies if this is wrong.

  • Comment number 58.

    #57. - newdwr54 wrote:
    "I based the '55 versions' notion on the fact that the previous UAH version was v5.4 and the latest is v5.5, so it seemed to me that they just go up in incremental steps? Apologies if this is wrong."

    On the contrary, I think it is I who probably needs to apologise.
    I was convinced that I had read that it was changing from v5 to v6, but I see it's from v5.4 to v5.5.
    However, since it's going from v5.5 to v6, I don't think it's 55 versions, obviously somewhere between 5 and 55!

  • Comment number 59.

    "Pacific eases further away from El Niño thresholds"

    Latest Enso Wrap-Up

    http://www.bom.gov.au/climate/enso/

    Not clear cut though, a sub sea cool east and warm west divide is developing. If the trade winds continue to slow the warm west could move to the east.

    http://www.bom.gov.au/cgi-bin/oceanography/wrap_ocean_analysis.pl?id=IDYOC007&year=2012&month=10

  • Comment number 60.

    Greensand . . well I'm confused.
    According to one link, Hadcrut4 is designed to better represent the global temperature record through the use of more arctic data - the inclusion of which, leads to a comparative warming of the more recent record. Yet as you point out, they've dropped a whole host of northerly canadian stations, which had been used previously.
    You would have thought that since the warming 'signal' is strongest at northerly latitudes, the arctic data might have been added to the existing canadian station data. Instead it's almost like they've replaced them.

    As you say, there may well be a good reason for this - perhaps it's a business/contractual thing?

  • Comment number 61.

    @60 lateintheday wrote:

    "Greensand . . well I'm confused."

    Sorry LITD probably me that is causing the confusion. In their justification for HadCRUT4 the MO did say that, amongst other issues, there was a lack of Arctic/Canadian data. They stated that they had data in that area in the 90's but not "now". "Now" being 2005-2010, prior to HadCRUT4, which would re-introduce the Arctic/Canadian data.

    My question was why did we not have the data in the 2005-2010 period? If the exclusion during the 2005-2010 period now partly justifies a whole lot of work to produce HadCRUT4 why were the stations removed in the first place? Also how is it possible to know how consistant a database is if the geographical input is changeable? Yes stations will come and go, or be off-line for a period and a gridded product will cope well. But a whole bunch of Canadian/Arctic data? The inclusion/exclusion appears to have come about via a decision process? Probably a damn good reason but it is not obvious.

    Also not sure if HadCRUT3 has the Canadian/Arctic data included after 2010? But not going there as HadCRUT3 will fade away. Always concerns me that the owners of the various databases both terrestrial and satellite take great solace in producing numbers that very closely match each other. Has not always been the case Phil Jones used to champion his HadCRUT numbers over GISS.

    History shows that most people benefit from having their suppliers in competition.

  • Comment number 62.

    The global HadSST2 anomaly for September is 0.453c, compared with 0.440c for August.
    N.H. is 0.608c, compared with 0.589c and S.H. is 0.298c, compared with 0.290c.
    I haven't looked at the precise figures yet, but this suggests a small rise in global and N.H. HadCRUT3 and very little change in the S.H. Otherwise not such large increases as those suggested by UAH & RSS. However, this is uncertain, given the influence of land temperatures in the N.H.
    HadCRUT4 uses HadSST3 data but I haven't been able to locate the monthly time-series data files for that dataset yet.
    greensand, do you know a location of the HadSST3 monthly time series in plain text format?
    The only files I can find are here, but they are in compressed format and none of them seem to be equivalent to the data files available for HadSST2.
    http://www.metoffice.gov.uk/hadobs/hadsst3/data/download.html

  • Comment number 63.

    Comparing HadCRUT4 with UAH v5.5 the 30 year trend is identical in both, to the 1/1000th of a degree (both +0.165C per decade).

    Over the past 10 years UAH v5.5 is running warmer than HadCRUT4 by +0.06C per decade (UAH shows +0.02 warming and HadCRUT4 shows -0.04C cooling in the past 360 months).

    Overall there is very good agreement among all the data sets, especially over the longer (30 year) period. In the last 30 years global temperatures have risen at an average rate of about +0.16C per decade.

  • Comment number 64.

    Hi Guys,

    Not seen this reported on the BBC, I wonder why?

    From Nature: Extreme weather

    Better models are needed before exceptional events can be reliably linked to global warming.

    http://www.nature.com/news/extreme-weather-1.11428

  • Comment number 65.

    #63. - newdwr54 wrote:
    "Over the past 10 years UAH v5.5 is running warmer than HadCRUT4 by +0.06C per decade (UAH shows +0.02 warming and HadCRUT4 shows -0.04C cooling in the past 360 months). "
    Is the UAH trend to September? I believe that to August it was only 0.011c/decade, which is strictly comparible with the HadCRUT4 figure.

  • Comment number 66.

    #64. - oldgifford wrote:
    "Better models are needed before exceptional events can be reliably linked to global warming."
    I read that to mean that it is assumed that only the deficiencies in the models is preventing them proving that exceptional events are linked to climate change.
    What they really mean is that "eventually we will create models which prove that such events are linked to "climate change".
    Actually I don't think that any model will ever prove that.

  • Comment number 67.

    65. QuaesoVeritas:

    Yes, the UAH rolling 10 year trend (+0.02) includes September, while the HadCRUT4 trend (-0.04) is to August. If HadCRUT4 reaches +0.55C or above in September then I think it will bring the 10 year trend up to -0.03C/decade? It's still quite a bit cooler than UAH v5.5.

 

BBC © 2014 The BBC is not responsible for the content of external sites. Read more.

This page is best viewed in an up-to-date web browser with style sheets (CSS) enabled. While you will be able to view the content of this page in your current browser, you will not be able to get the full visual experience. Please consider upgrading your browser software or enabling style sheets (CSS) if you are able to do so.