« Previous | Main | Next »

All eyes on new october temperature records

Paul Hudson | 13:20 UK time, Friday, 30 September 2011


New records were set on saturday.

Bramham now holds the Yorkshire October record, with 28.7C. Finningley recorded 30C, but the Met Office do not regard the airport as an official climate site anymore.

A new record was set in Lincolnshire on sunday, with 29.3C recorded at Coningsby.


The exceptional early autumn weather looks set to last into the weekend, with UK temperature records on the line.

According to the Met Office the current UK October record was set on the 1st of October 1985 in March, Cambridgeshire, where 29.4C (85F) was observed.

Closer to home, 1st October 1985 also saw a new record in Lincolnshire, with 28.4C (83F) at Cranwell.

In Yorkshire, the highest October temperature on record was set in Whitby on the 2nd October, 1908, with 28.3C (82F).

Tomorrow both Lincolnshire and Yorkshire could see new October records, with 29C (84F) possible in one or two spots.

Already at 2pm today, the mercury at Finningley in South Yorkshire has hit 29C, with very similar meteorological conditions expected tomorrow.

The Met Office estimate a 30% chance of 30C (86F) being recorded at one of their stations across Eastern or Southeastern England.

A fall in temperatures is expected across all areas early next week.

By the second half of next week, night-time temperatures could fall low enough for a touch of ground frost in some sheltered locations, as weather conditions return to normal for the time of the year.


  • Comment number 1.

    I have just been reading the article "Flowers bloom for a second time this year"
    by Jennifer Carpenter in which she says Wakehurst Place's head Andy Jackson said. "In the last year, the UK experienced an incredibly warm winter"

    I thought it was the 18th coldest winter since 1910?

  • Comment number 2.

    It has been another dry month, though, if a windy one. Here in Staveley, between Sheffield and Chesterfield I have only recorded 33.7mm of rain, making just 310mm for the year to date. We have just travelled from Doncaster and they are irrigating the fields near Armthorpe. A bit of warmth at this time of year is nice, though.

  • Comment number 3.

    It seems that the article has been a little "economical with the truth", in order to make a story. I bet a bit of digging into the "facts", might reveal that some of these events are not as unusual as they would appear.
    I am not a botanist, but my old "Observers book of Wild Flowes", states that the Foxglove flowers from June to September, so nothing apparently unusual there.
    Also, some varieties of Buddleia flower from July to September. Are these really flowering twice or simply flowering late? Is there actually any proof that the same plant has flowered twice (By the way, these are all rhetorical questions)? My own Buddleia flowered some time ago and shows no sign of flowering again. Is it even physically possible.
    The Cowslip does normally flower in April & May, so any flowering now would be unusual, but I would want positive evidence.
    As for an "incredibly warm winter", this is again misleading. December 2010 was the second coldest in the CET record, whereas February was the 22nd warmest, but I wouldn't call that "incredible", since it wasn't as warm as 1779, 1794, and several other years in the 18th century. Even in the South East of England, the February mean of 6.5c, was above average, but below that for 2002, 1998, 1995 and 1990.
    Since this "warm spell" has only lasted a few days, prior to which, temperatures have been about normal, I don't really see why plants would have suddenly started to behave as if it is spring!

  • Comment number 4.

    The flowers shown in the BBC video piece were flowers that appear in the late winter into spring. The very warm winter seems like a strange error, it was a warm (and very, very dry) spring round here, but not winter. It seems to me the article was written by someone who knows nowt about gardening and nowt about weather and was checked by an editor while (s)he was staring out of the window.

  • Comment number 5.

    So has there been an item about this on the BBC News?
    Must set my recorder to see if I can catch it.
    If the flowers normally appear in late winter/spring, does that not suggest
    something odd?
    Maybe I need to see the video to understand the context.

  • Comment number 6.

    Found this while searching for the flowers article.
    I don't really understand what is meant by "social aspects".

  • Comment number 7.

    Seems the UK's '30th September high temperature' record just went: http://www.google.com/hostednews/ukpress/article/ALeqM5gLmg2F5fhQNhBRj8X_kEri7zJedg?docId=N0678841317375632185A

    Oddly this is happening at a time when global surface temperatures are plummeting, possibly due to La Nina conditions stripping heat out of the atmosphere and into the ocean in the Pacific.

  • Comment number 8.

    Will tomorrow, Saturday 1st October 2011, be the warmest October day on record in the UK?

    I'd rate it as a 6/10 chance. Anybody got any less cautious predictions? (Preferably *before* noon on 1st October, thank you.)

  • Comment number 9.

    I think it is slightly odd that has taken over 100 years to break this record.
    We know that global annual temperatures are about 1c warmer now than in 1908, (based on HadCRUT3), so in that respect, it isn't surprising that given the right conditions, record temperatures have been set.
    In this case, we know specifically why we are getting high temperatures, i.e. the blocking high over Europe, and the warm Southerly winds. So really it's difficult to say if the precise level is due to "global warming", or sheer chance.
    I don't know if weather charts exist for September 30th 1908, but presumably if they do, they would show a similar weather pattern to that which currently exists.

  • Comment number 10.

    Trouble with these forecasts, then the reports and comments is the almost fanatical expectation of something record-breaking. Incredible warmth, exceptionally dry or wet or warm or cold . . are all adjectives over used by media, scientists and now the public - who aren't interested in a spot of decent weather for what it is unless it's 'smashed' a record somewhere.

    MY first week at work in the Met Office at the end of a Sept (many years ago) was very warm, followed by a reasonable October, stormy Nov and a cold spell with snow at Christmas. Those were the days when, v cold, cold, normal, mild, warm, v. warm, hot all had a numeric definition as to when they should be used in Met forecasts. It was also frowned upon - especially in Met O 6 (Military) to use terms such as sleet! It had to b described as rain and snow.

    I know normality doesn't sell media space but wouldn't it be nice to get back to objective reporting, commenting etc . . . at least on this site? I write this with observations (00Z 1 Oct) showing surface temps of 17 to 20 over Yorkshire and a slack surface gradient with Paul's 564 line about over Yorkshire for 12Z. Whilst I couldn't comment on it being the warmest ever Oct day in the UK (it is more likely to be earlier than later in the month) I would not be surprised if local extremes are 'beaten'!

  • Comment number 11.

    QV I note that you have unearthed the up to date HadCRUT3 temperature
    Your moving annual figure suggests that the Met.Office have yet again overestimated annual temperatures

    Year - Prediction - Reported outcome
    1999 …0.38…..…0.26
    2000……0.41……… 0.24
    2001……0.47……… 0.40
    2006… 0.45..……0.43


    My forcast for next year is Met.Office minus 0.08

  • Comment number 12.


    The official MO HadCRUT3 for 2011 to the end of August, is now 0.372c, compared to 0.349c at the end of July, i.e. an increase of 0.023c, despite virtually no change in the monthly anomaly figure since last month. This compares with an increase in the simple average of the monthly figures, from 0.342c to 0.357c.
    As it is not possible to replicate the MO method of annual anomaly calculation, it wouldn't surprise me if the final figure turned out to be quite close
    to their original prediction of 0.44c for the year.
    The problem with the MO method of annual anomaly calculation, is that while it may strictly speaking be more accurate than the CRU method, it is also less transparent, and considerably more difficult for an "outsider", with limited resources, to replicate.

    The HadCRUT3 hemispheric anomalies for August were as follows:

    NH = 0.620c (July = 0.595c)
    SH = 0.296c (July = 0.333c)
    My own estimates were 0.559c +/- 1c and 0.273c +/- 0.5c, so whereas based on HadSST2, I was expecting a small fall in both hemispheres, there was
    actually a small rise in the NH figure. It is interesting that if I had based my estimates simply on last months changes in the HadSST2 anomalies,
    I would have obtained more accurate results than from using a more complex relationship, so possibly "back to the drawing board" on this one.

  • Comment number 13.

    After rising slightly between September 21st and 22nd, the AQUA CH5 temperature has resumed it's downward path, slightly faster than normal for the time of year, but not as quickly as between September 15th and 17th.
    As a result, the cumulative temperature for the month is still pointing towards a UAH figure of about 0.4c, an RSS figure of about 0.37c and a HadCRUT3 figure of about 0.46c for September. Actually the theoretical UAH figure is strictly speaking about 0.3c, but during the last few months, the actual figure have tended to be about 0.1c above the theoretical, based on the period 2003-2010.

    Taking August anomalies into account, the above would seem to suggest a rising trend of temperatures, which might continue into October, but that is slightly misleading.
    Most of the increase in temperature occurred during the first half of September and during the second half, temperatures have actually been falling rapidly.
    During early September, I expecting the final UAH figure for the month to be much higher than 0.4c.

    Breaking the month of September into weeks, and calculating equivalent UAH anomalies, the results are as follows:

    Sept. 1-7 = 0.434c
    Sept. 8-14 = 0.501c
    Sept. 15-21 = 0.359c
    Sept. 22-28* = 0.337c

    Consequently, if temperatures continue to fall in a similar manner to those during late September, it is possible that the October UAH figure will return to levels similar to those in August. Of course, it is also possible that temperatures will not
    fall as quickly as expected during October. The above figures include a 0.1c "adjustment", to allow for the fact that recently, UAH figures have been higher than theoretical estimates based on the period 2003-2010.

    * AQUA CH5 temperature figures are not yet available after September 28th.

  • Comment number 14.

    "Oddly this is happening at a time when global surface temperatures are plummeting, possibly due to La Nina conditions stripping heat out of the atmosphere and into the ocean in the Pacific."

    Warming something up makes it cooler? Missed that one at school.

  • Comment number 15.

  • Comment number 16.

    15. mjmwhite:

    The ocean surface temperature anomaly is not a reliable guide to ocean heat content. It's just the top mm of ocean surface.

  • Comment number 17.

    9. QuaesoVeritas:

    I agree that the UK temperature record is generally a poor proxy indicator for global climate patterns. No doubt the recent warm temperatures are a weather phenomenon.

    On a global scale however, it's interesting to note that the max temperatures recorded for each month in the Hadley record all occurred during the past 15 years; whereas all the mins occurred before 1918, with most of them being over 100 years old:

    Global monthly temperature max/min (HadCRUT3):


    Jan 1893/2007
    Feb 1862/1998
    Mar 1917/2002
    Apr 1911/1998
    May 1861/1998
    Jun 1903/1998
    Jul 1909/1998
    Aug 1862/1998
    Sep 1859/2003
    Oct 1864/2003
    Nov 1862/2004
    Dec 1892/2007

    In my view this pattern is unlikely to have occurred by 'sheer chance'. I don't really see how this apparent increase in overall heat energy can be explained by natural cyclical patterns of warming and cooling either.

  • Comment number 18.

    17. Amendment @ 16:

    table should show:


    'Dec 1892/1997'

    Not '1892/2007'


  • Comment number 19.

    "The ocean surface temperature anomaly is not a reliable guide to ocean heat content. It's just the top mm of ocean surface."

    "Understanding Sea Surface Temperature"


    Slightly more complicated than 1mm.

  • Comment number 20.


    AS you can see to the east of Leeds we have had 230 mm of rain this year.. That is under 10 inches.

    Thats the lowest i can find EVER after 9 full months.
    The rain we have had has fallen in small amounts and has been removed by wind and evaporation.

    We are so dry out here in the vale of york, that no-one can give me any stats that are even close.
    Anyone in the area have a look at the area around Fairbun ings to see how dry we are.
    Many of the smaller ponds are dry and unless we get a river breach are likley to remain so.

    i am going for a wet and as such warm winter overall, to re balance rainfall. That said a serious cold snap could and will occur at some stage.


  • Comment number 21.

    "due to La Nina conditions stripping heat out of the atmosphere and into the ocean in the Pacific"

    2 problems,

    1 During a period of La Niña, the sea surface temperature across the equatorial Eastern Central Pacific Ocean will be lower than normal. The statement seems to indicate that heating the ocean makes it cooler.

    2 A warm ocean would heat the atmosphere not the otherway round. You don't use a hair dryer to heat your bath.

  • Comment number 22.

    19. mjmwhite:

    With respect you were referring to NOAA sea surface data which uses the top millimeter of the ocean’s surface as measured by satellites. However even if you want to use the top 20m as 'surface' data, this is is still a very poor indication overall ocean heat content as it is part of the 'mixed layer'.

  • Comment number 23.

    21. mjmwhite wrote:

    "1 During a period of La Niña, the sea surface temperature across the equatorial Eastern Central Pacific Ocean will be lower than normal. The statement seems to indicate that heating the ocean makes it cooler."

    During La Niña conditions warm water in the upper mixed layer of the ocean is blown west from the Eastern Central Pacific, where it is replaced by up-welling cold water. The heat is not lost from the ocean; it's just that a cooler layer is allowed to rise up from beneath. It's this cold water that has a cooling impact on global temperatures. The ocean has not suffered a net loss in heat energy.

    "2 A warm ocean would heat the atmosphere not the otherway round. You don't use a hair dryer to heat your bath."

    That's right. But the ocean is stratified due to both temperature and salinity. If a seam of cold water rises to the surface, as happens during La Niña, then it cools the air at the ocean/atmosphere interface. But this doesn't say anything about the ocean's 'overall' heat content, which appears to be increasing right down to 1,500m: http://www.skepticalscience.com/Ocean-Heat-Content-And-The-Importance-Of-The-Deep-Ocean.html

  • Comment number 24.

    Paul, can you say something about the criteria used to measure the temperatures that we see reported " x degrees in Gravesend" or whatever.

    I live on the outskirts of Leeds and my thermometer in the garden shade has read over 30 for today and yesterday. I'm sure it's even hotter in the city centre. So why not "Leeds sees 30 degrees" for the headline ?

  • Comment number 25.

    with respect, 'stripping heat out of the atmosphere' was hardly the most accurate turn of phrase.
    During La Nina, the atmosphere is warmed less by the ocean. It is not cooled. All things being equal, this should mean that OHC goes up.

    The apparent 'discovery' of a rise in OHC below 700m is extremely weak. Some have pointed to the fact that there had been no indication thus far of such an increase, and no sign of vertical heat transfer despite direct observations.

    Be very careful about adopting the 'sinking at the poles' explanation. This cannot be invoked without equal consideration that the thermohaline circulation which is believed to be in the order of 1200 - 1800yrs, may be responsible for the observed 20thC warming - and as such, perfectly natural.

  • Comment number 26.

    Well it appears the record has been broken - 29.9C so far and a possible 30C somewhere amongst other stations that will report after 7PM. Not bad, at least hopefully it will make the winter seem a bit shorter

  • Comment number 27.

    25. lateintheday:

    "with respect, 'stripping heat out of the atmosphere' was hardly the most accurate turn of phrase"

    Yes, you're probably right. Saying 'La Niña conditions warm the planet less than ENSO neutral conditions' is a better way of putting it. In any case, the global surface temperature anomaly tends to fall relative to 'normal' during La Niña conditions.

    "The apparent 'discovery' of a rise in OHC below 700m is extremely weak."

    Admittedly it's uncertain. But the Palmer 2011 study [from SkS link] suggests that the 'missing heat' may be accumulating in the oceans below 700m. I don't find this to be a stretch, given that the oceans are undoubtedly earth's primary heat store. Why should heat gained at the surface stop accumulating below 700m?

    "....thermohaline circulation which is believed to be in the order of 1200 - 1800yrs, may be responsible for the observed 20thC warming - and as such, perfectly natural"

    It may be a natural cycle. We don't have any accurate way of confirming what temperatures on earth were like 1200-1800 years ago. It may just be a coincidence that the recent observed warming has occurred during a period when intensified greenhouse gas concentrations have been measured in the atmosphere.

  • Comment number 28.

    October may have a warm couple of days, may even break the 'record'.

    The global average temperatures are in for months up to August, though HadCrut3 only to July, and all show 2011 cooler than previous years going back to 1998. The UK is a small island off the NW coast of Europe and a record couple of days heat will not affect global averages. The complete list as available on the wattsupwiththat web site and easily accessed.

    Some bloggers might not trust this site though I find it completely trustworthy.

  • Comment number 29.

    John Marshall wrote: "Some bloggers might not trust this site though I find it completely trustworthy."

    I think that says a lot more about the standard of credibility you accept than it does about the inaccuracies in WTFUWT. For example;


    Canadian Harp Seals are psychic, or, have evolved precognition? Completely trustworthy? Really?

  • Comment number 30.

    28. John Marshall wrote:

    "The global average temperatures are in for months up to August, though HadCrut3 only to July, and all show 2011 cooler than previous years going back to 1998."

    HadCRUT3 temperatures have now been published up to the end of August and are available here from the CRU itself: http://www.cru.uea.ac.uk/cru/data/temperature/hadcrut3gl.txt

    They show that 2011 Jan-Aug is warmer than were Jan-Aug 1999, 2000 and 2008. Anthony Watts must have got confused.

  • Comment number 31.

    Wow! AMSU temperatures continue to drop into October
    It is now below all years from 2002 to 2010 EXCEPT 2004

  • Comment number 32.

    Yes, I think it might have dropped below average on the 30th for the first time since May 26th.
    On a cumulative basis, it is now below all all years (since 2002) except 2008.
    If this keeps up, it should show up in a low October anomaly.

  • Comment number 33.

    #26 - millennia wrote:
    "Well it appears the record has been broken - 29.9C so far and a possible 30C somewhere amongst other stations that will report after 7PM. Not bad, at least hopefully it will make the winter seem a bit shorter."
    Apparently the 29.9c record for Gravesend has been cancelled because Gravesend doesn't conform to MO regulations, which seems like a bit of an oversight.
    The next highest was 29.4c at Charlwood.
    UKMO still haven't changed their press release however.

  • Comment number 34.

    To QV#3 & kelvdfl #1

    Been away in Cumbria over WE so a bit behind (incidentally was mostly dull and unexciting weatherwise there)

    These "flower out of season" stories are always popping up and in my view rarely add up to much. It is virtually normal for many spring flowering shrubs to have a few (sometimes quite a lot of) flowers in the autumn. Unusual rainfall patterns can also trigger unseasonal blooming (particularly summer drought). But I think a large number of reports are down to lack of observation at other times. People tend to notice flowers on a warm sunny day - so when the weather is hot - they suddenly see (and indeed probably expect to see) all sorts. Of course no year is ever the same. You also have to allow for horticultural variety. There is actually a daffodil called 'January' as well as forms of Crocus and snowdrop that bloom in autumn and winter quite naturally.

    Once in the press this contributes to the notion that "the seasons are all over the place". You only have to read old weather diaries (like that of Gilbert White - 18th cent) to realise that they had some pretty unseaonable weather then.(Eg pond ice in June!)

    The response of plants to temperature can be quite complex even to the extent that a very mild winter will not necessarily produce the earliest flowers in every case. That does not mean that "climate change" is not having an affect on plants - only that one has to be very cautious about jumping to conclusions.

    Another topic - rainfall

    Some springs (water!)are dry here that I can never remember being dry before. Despite occasional dampening by fairly decent rain, the moisture obviously hasn't gone in far. Of course one also has to allow for water abstraction affecting springs. In this area (porous chalk) a lot of water is taken out and can influence water table considerably. Even so levels of chalk spring water seem unusually low now.

  • Comment number 35.

    QV @ 33 & 32
    Finningley recorded 30C twice I believe, 30th Sept and 1 Oct, but again it isn't an official station. I'm surprised at Gravesend because it often records the highest temperatures in heat waves (perhaps suspect in itself, because it is on the coast) and I thought held the "above 100F" record for the UK.

    On the subject of anomalies I see big Joe B is back on the warpath over collapsing temperatures again and has a prediction out for a low of -0.2C by March. So it's eyes on the predictions we all made at the beginning of the year again. I've pretty much given up on it dropping back to my +0.24 (MO comparative scale) but reckon I'll be a heck of a lot closer than the MO themselves, who were listening too much to "Super El Nino" Hansen again when we have actually have the double La Nina forming that JB was indicating.

  • Comment number 36.

    I noticed you have quoted HadCRUT3 recorded global maxima and minima
    I decided to look back through solar cycle records
    I have quoted the solar maximum at the time of each low temperature year quoted
    I have only eyeballed the solar data
    If the low temperature year occurred at a solar min. I have quoted the previous cycle max.
    Nearly all low temperature years occurred at times when the solar max was less than 90

    From 1998 to 2007 the solar max was between 120 and 150


    Jan 1893 (60)/2007
    Feb 1862 (95)/1998
    Mar 1917 (100)/2002
    Apr 1911 (60)/1998
    May 1861 (90)/1998
    Jun 1903 (80)/1998
    Jul 1909 (60)/1998
    Aug 1862 (95)/1998
    Sep 1859 (120)/2003
    Oct 1864 (90)/2003
    Nov 1862 (90)/2004
    Dec 1892 (60)/2007

    I agree that this pattern is unlikely to have occurred by 'sheer chance', but I do think that natural SOLAR cyclical patterns may have something to do with the global trends you have noted.

    The next two solar cycles are likely to have solar peaks around 65.
    I for one will be waiting to see what happens to global temperatures over the next 25 years before dismising natural effects on global temperatures

  • Comment number 37.

    Look like Gravesend suffers too much from UHI ;) http://t.co/mAd1qr1S

  • Comment number 38.

    I'm not sure what newdwr54 is trying to say here. There is little doubt or argument that on the whole, the world has progressively warmed since the little ice age. It seems somewhat tautological to state that the coldest average months occurred when the world was cooler. Similarly, one would expect the warmest average months to occur when the world is at its warmest in the measured record. I should expect that a new record warm year may be set when we get the next big el nino, a year or so after the upcoming solar max - even if that max is sub 90 ssn. This in itself, would not worry me unduly since it would represent maximum cycle forcing on top of historically high OHC.

    Perhaps newdwr54 is simply addressing those few who don't believe that the world has actually warmed during this period. I think it's fair to say that the majority of skeptics question whether the rate of the late 20th C is as unusual or unnatural as the consensus portray. Whilst many question the accuracy of the recorded temps, their issue is really around whether bias has crept into the marginal measurements.

    We're still waiting for BEST to publish fully. As I understand it, no one is seriously expecting their results to be out of step with the usual suspects. Skeptics like myself, would be pleasantly surprised if they returned results which were significantly below the combined average of main players.

  • Comment number 39.


    I am not entirely sure how you have arrived at the solar max figures you quote, but it is true that 50% of the monthly minimums were set during the period 1879 and 1934, when solar maximums barely rose over 100. All of the other minimums were set between 1861 and 1864, and the sunspot maximum of 1860 was relatively low, compared to those around 1837, 1848 and 1871. Of course, the HadCRUT3 record only begins in 1850, but I think it may be significant that there were apparently no record lows set between 1850 and 1860.

    On the other hand of course, the period of warming which has resulted in the recent high temperatures, has occurred when all sunspot maximums have been above 100 and several above 150. Indeed, by my calculations, the three highest sunspot maxima have occurred since 1958. Also, the 100 year mean sunspot number peaked in 2004 and is almost as high now. The 20 and 50 year averages were both at very high levels during the 1990's.

    Now, is it just coincidence that the period of warmest global temperatures has occurred during a period when both CO2 levels and sunspot activity have been at their
    highest? It would be interesting to compare the correlation of sunspot activity and temperature, versus CO2 and temperature, but unfortunately, since we only have CO2 figures going back to about 1960, since when sunspot activity has been relatively high, I don't think that would tell us a great deal.

    However, if as predicted sunspot activity is about to enter a period of decline, while CO2 levels are continuing to rise, we may learn something over the next few years.
    At the very least, I think this suggests that there is at least one other factor to be taken into account when searching for the cause of global temperature rise over the last 100 years or so.

  • Comment number 40.

    "Indeed, by my calculations, the three highest sunspot maxima have occurred since 1958. Also, the 100 year mean sunspot number peaked in 2004 and is almost as high now."

    I thought it was generally accepted that 4 of the last 5 solar cycles had been the most active on record but that the peak had occurred circa 1960. I remember you saying some months ago that you had not really examined the solar record to any great extent in the past, but that you were minded to. One thing you may have missed in your research is that for some reason (according to Leif Svalgaard) current SSN need to be multiplied by 0.6 in order to find the monthly smoothed average. This was pointed out to some chap who had noticed that a recent spurt in solar activity had taken daily SSN to over 100.

    And sorry, but what exactly is a '100year mean sunspot number' ?

  • Comment number 41.


    I don't know how it is done "officially", but I use the 12 month MA SSN to estimate the peak of each cycle. Based on that, I make the cycles peaking in 1958, 1980 and 1989 the most active, although that may not be the autodox method. I make the peak in 1778 slightly higher than the one in 1947.
    When you say that the "current SSN" need to be multiplied by 0.6, do you mean the daily ones which are published on sites such as "Spaceweather"?
    I noticed some time ago that they tended to be higher than the official monthly ones, which I normally use. Actually in order to convert the daily to monthly figures, I multiply by 0.7439 and deduct 4.3114. Is there a reason for the figure of 0.6?
    By a 100 year mean sunspot number, I mean a rolling 100 year (or 1200 month) average of the monthly figures.
    Are you saying that the official monthly figures are smoothed?
    If so, I must have missed that. I don't see why smoothing them would make them a constant 0.6 of the daily figures, assuming that is what you mean.

  • Comment number 42.

    36. NeilHamp:

    "Nearly all low temperature years occurred at times when the solar max was less than 90"

    Thanks for that. That's very interesting.

    As I understand it, the recent CERN/CLOUD study was examining the impact of cloud formation on climate. There is a theory that during periods of low sunspot activity, which leads to lower solar output, galactic cosmic rays get the chance to seed more clouds.

    The theory then goes (I think) that greater cloud formation leads to increased warming, especially at night, since clouds trap IR in the atmosphere. This would neatly explain why warming is occurring, also why more warming is occurring at night.

    The problem is that no one knows for sure what the net impact of clouds is on climate. The IPCC says it's slightly negative. Not only that, but as you have shown, temperatures are usually observed to fall during periods of reduced solar input - not to rise, as the 'cloud' theory suggests.

    It seems rational enough that global temperatures should fall rather than rise during periods of reduced solar output. And the evidence appears to support this view. However it is the *opposite* of what theories such as those of Svensmark expect, if I have understood them properly.

    Does it not also leave us in a position where we require an explanation for why the measured increase in greenhouse gases has *not* influenced global climate to the expected extent?

  • Comment number 43.

    38. lateintheday wrote:

    "There is little doubt or argument that on the whole, the world has progressively warmed since the little ice age."

    I agree with that. But it still seems remarkable to me that all the warmest monthly temperatures on record have occurred within the past 15 years.

    Furthermore it does not address the question of what has caused this warmth.

  • Comment number 44.

    QV - I'm not the person to ask about this and I've probably misunderstood Leif's comment. I would suggest dropping any questions to Tallbloke's Talkshop. He's responded very politely to a few (very basic) questions I have posed in the past.

    The x 0.6 figure seems to be the standard calculation to correct for differences in telescope refractors since the time of Wolf and I think is applied to daily counts. More detail from this SIDC link, under section 5, Calculating the International Sunspot Number.


    By the way, did you choose the 100yr moving average for any particular reason? Tallbloke has posted a few times on Periodicity. Apparently there are signals beyond the usual 11 and 22yr cycles and I seem to remember one being around 80yrs.

  • Comment number 45.

    At 21:55 3rd Oct 2011, newdwr54 wrote:

    'The theory then goes (I think) that greater cloud formation leads to increased warming, especially at night, since clouds trap IR in the atmosphere. This would neatly explain why warming is occurring, also why more warming is occurring at night.'

    Not quite there:

    The rationale for this is the following: if cloudiness is high in the daytime,
    more sunlight is reflected back to space and the daily temperature maximum is lowered; in the nighttime, less infrared radiation from the earth surface is emitted into outer space and the daily temperature minimum is increased. Therefore - more clouds means lower DTR.

    Further info:

    I seem to recall a date of 2007 (end of cycle 23?) as being indicative for the start of the solar minimum which in theory should increase this effect, perhaps someone could look at min and max temps for the current period to see if there is anything noticeable, possibly need to have a decade or more of data though?

  • Comment number 46.

    Thanks for the link, I will follow that up.
    No particular reason for choosing 100 years, it was just a long period over which
    to monitor the trend. I also calculated 50 year and 20 year averages. While the average cycle period is 11 years, the actual period varies, so in 100 years there would be 9 or 10 cycles.

  • Comment number 47.

    you've probably seen this before, but nevertheless it's worth reviewing. This link shows a very interesting 'coincidence' between falling tropical cloud cover and rising global atmospheric temps.


    This may be a pointer that clouds have a net negative feedback around the tropics. It may well be that they have a net positive feedback at higher latitudes. Quantifying the net overall feedback is someone else's headache.

    The lack of good correlation between sun/earth temps post 1960 may be partly explained by this drop in cloud cover, since as we know, TSI does not vary greatly within cycles. The cause of the drop in cloud cover? I'd like to think that this could be linked to variations in the solar UV spectrum (which we know is much greater than TSI) but in all honesty, this is just wishful thinking for now. I'm rather taken with some of Stephen Wilde's ideas though, which takes some of the science at the molecular level and seeks to predict how this might manifest itself in weather patterns.

  • Comment number 48.

    "Furthermore it does not address the question of what has caused this warmth."

    You're quite right. However, with the timescale that we're talking about, (few hundred years) I thought that the AGW explanation had some problems of its own. Something along the lines of . . . the modelled CO2 forcing required to support late 20thC warming cannot explain the early 20thC warming. Thus more of the earlier warming is attributed to natural variation.

    Maybe this is just a blog myth that gets passed around.

  • Comment number 49.

    I fear that the discussion above regarding clouds does not reflect the full complexity of the issue. The fall in cloud cover from 1983 to 2001 was associated with a parallel fall in aerosols. The impact of this has been discussed in detail by Pinker et al (2005).

    Pinker et al concluded that the net radiative forcing from "global brightening" was actually very small because the increased solar radiation was almost completely cancelled out by decreased absorption of IR radiation.

    Meanwhile, both daytime and nighttime temperatures have increased significantly, although nighttime temperatures have increased more, reducing the diurnal temperature range. This is consistent with an enhanced greenhouse effect but not with reduced cloud cover.


  • Comment number 50.

    there does not appear to be much in the way of unusual heat energy in the earth system according to the latest figures from Roy Spencer


    smoke me a kipper

  • Comment number 51.

    lateintheday @ #47

    I've now had a little time to study the cloud cover graphs at climate4you.

    I'm sure you're aware that correlation does not necessarily equate to causation. In this case, for the reasons stated above, it is likely that the reduction in cloud cover from 1983 to 2001 was only responsible for a small proportion of the warming trend over the same period.

    I'm also concerned that climate4you's estimate of a 0.07C temperature decrease for a 1% increase in cloud cover may be flawed. Perhaps I've misunderstood what they have done, but it appears that they are assuming that the spread of points around the regression represents the other factors associated with the warming trend. In reality though, the spread of points is almost certainly down to natural variation in global temperature due to factors such as ENSO and solar cycles - ie. factors with no discernible trend over the period in question.

    Over the same period, there has also been a steady upward trend in CO2. I can't see how the method used by climate4you can hope to differentiate between any effect of clouds and the effects of CO2.


  • Comment number 52.

    That's interesting.
    The "adjustment" factor of 0.1c, which I have been adding to my UAH prediction based on AQUA CH5, seems to have disappeared.
    I said in my post 13 that the theoretical figure was between 0.3c and 0.4c, and the actual, 0.289c, is close to the lower figure.
    So, for whatever reason, the UAH anomaly is back where it should be, based on AQUA CH5.
    Nobody should expect a similar fall in other anomalies, e.g. RSS and HadCRUT3 as a result of this. In fact, the opposite is likely.

  • Comment number 53.

    Paul B, thanks for that - I'll take a closer look.
    I've tried to find a good debate around Pinker et al 2005 but have yet to find one. My first visit was to SkS thinking that firstly they would explain the science, secondly they would present the findings in a pro AGW light so that inevitably, I'd find some alternative viewpoints/debate in the comments section. By reading both sides, I get a better overview.
    Their coverage is disappointing on this however. It reads as though the author (Cook?) started off with all good intent and then ran out of time and simply threw in a GISS graph to finish. Oddly, despite this being a 2010 post, there were no comments - none at all.

  • Comment number 54.

    The October 1908 heatwave in Yorkshire appears to have been even more remarkable than the one we have just experienced. The data I have seen from UKmetworld (which lists highest temperatures for each day) suggests that the peak of the heat was in Whitby, which had four days in excess of 28C with 28.9c on 1 October, 28.1C on the 2nd and 28.3C on both the 3rd and 4th. Incidentally the 28.9C on 1st would still be the Yorkshire record (not sure whether someone else can check this???).

    Interestingly only 2 years earlier - in 1906 - there was a remarkable autumn heatwave at the start of September, with 35.6C (I think) recorded near Finningley and three of four consecutive days over 32C.

  • Comment number 55.

    50. Spanglerboy wrote:

    "there does not appear to be much in the way of unusual heat energy in the earth system according to the latest figures from Roy Spencer"

    Spencer's UAH data says September 2011 was +0.29 C above the 1981-2010 average.

    That moves 2011 year to date into 6th warmest on record according to UAH; this despite natural La Nina cooling that persisted until mid May and which started up again in mid September.

    September 2011 is also the sixth warmest September in the UAH record. Don't cancel 'global warming' just yet.

  • Comment number 56.

    Chris @54
    It is amazing that 2 of 3 years should contain such record and more importantly sustained heat at a time when global temperatures were considerably cooler than today, although beginning on a march upwards to the 1940s that was steeper than 1976-1998 after a further cool period coinciding with low sunspot numbers around 1913.
    Can you even begin to imagine the headlines if we had sustained 90F heat in September today?

  • Comment number 57.

    47. lateintheday:

    If tropical cloud cover has been falling since 1883 then I'd not be surprised if it's had an impact on climate increase, since you'd expect tropical SSTs to increase, which in turn would increase evaporation of water vapour, etc, etc.

    However I think we'd need to know a lot more about the nature of the clouds in question. For instance reduced 'total cloud cover' might mean a disproportionate decrease in the high cirrus clouds that are thought to increase warming. If the low, thick cumulous clouds that promote cooling and less evaporation persisted, then it would be much harder to support a strong role for tropical clouds in the observed warming (+0.16C per decade since 1983).

    48. lateintheday:

    ". . . the modelled CO2 forcing required to support late 20thC warming cannot explain the early 20thC warming. Thus more of the earlier warming is attributed to natural variation."

    That's the current position of the IPCC. The observed surface warming pattern of the 20th and early 21st centuries cannot be accurately modelled unless both natural and man made forcings are taken into account. The same goes with Arctic sea ice loss models btw.

  • Comment number 58.

    Do you mean the observed surface "cooling" pattern of the early 21st century?

  • Comment number 59.

    according to the climate4you site, the cloud cover decrease over 15N - 15S since 1983 shows that low level cloud decreased more than high level.

    I think that the correlation with global temps is interesting. Equally interesting is why the decrease in cloud cover occurred. As PB points out, according to Pinker at al, there was a similar decrease in aerosols. Since aerosols are thought to be strong candidates for heterogeneous nucleation, might it be possible that these two events are linked. Also, what happens to the water vapour that doesn't make it into cloud formation? Perhaps this would show up as a rise in humidity.

    too complicated - my head hurts.

  • Comment number 60.

    58. QuaesoVeritas wrote:

    "Do you mean the observed surface "cooling" pattern of the early 21st century?"

    Very good QV; should have seen that one coming.

    I was thinking more of the continuing current 30-year warming trend which is clear from all the global data sets. This has slowed of pace lately, but this levelling off has occurred at an exceptionally high 'plateaux'.

    On the wider point, there is no dispute over the fact that there are forcings on climate other than CO2. That's the main reason why the WMO recommend using a period of 30 years continuous temperature data to identify underlying climate trends from the noise.

  • Comment number 61.

    56. Millenia:

    Yes, I think too many people seem to think that extreme weather is a recent phenomenon whereas heatwaves and cold spells have always been irregular visitors to the UK. There were some very impressive autumn heatwaves in the the last decade of the 19th century (the 1890s) as well as in the first decade of the 20th. My previous post (no 54) pointed out the autumn heatwaves of October 1906 and September 1908, but there were impressive heatwaves in the LAST week of both September 1895 and 1898 with temperatures between 28-30C across much of the country. And also remember that the summer of 1911 arguably produced the highest recorded maximum temperatures in the UK until recent times, with peak temperatures exceeding 35C.

  • Comment number 62.

    naughty naughty newdwr54 @ 55

    i referred to the earth system. you speak only of the surface temperature which we all know has no scientific relevance. tale a look at SST and tell us what you see

    no Crackerjack pencil for you

    smoke me a kipper

  • Comment number 63.

    actually Spanglerboy, we don't all know . . or at least I don't.
    Can you explain this a bit more or point me to an article please?
    I think I'm even more confused today than normal, but I thought that the UAH record was a combination of atmospheric temp (CH5) and SSTs (AMSR-E) data.

    BTW I see that the latter is now defunct, but will be replaced in a year or so.


  • Comment number 64.

    lateintheday @ #59

    "Since aerosols are thought to be strong candidates for heterogeneous nucleation, might it be possible that these two events are linked."

    Yes, it is very likely that they are linked, at least to some extent.

    "too complicated - my head hurts."

    In truth, even the scientists researching this area don't claim to fully understand it!


  • Comment number 65.


    temperature is an intensive property and as such cannot be averaged. Global average surface temperture is used by politicians and advocates to scare the children.

    By way of a simple example it requires much less heat energy to raise the temperature of a given volume of cold dry air 1c than to raise the same volume of warm humid air 1c.

    That is not to say that CO2 is not a greenhouse gas. It is. Changing the amount of CO2 in the atmosphere may affect climate in the long term. However, in my view it is a bit player. Sun sea and clouds drive the climate. Everything else is just along for the ride.

    kind regards

    for you I will smoke my own kipper

  • Comment number 66.

    The RSS global anomaly for September is 0.288c, compared to last months figure of 0.291c for August, although that figure has been slightly revised to 0.286c.
    The September figure is equivalent to an anomaly of 0.435c, adjusted to the same time base as HadCRUT3.
    Although this figure has not fallen, as was the case with UAH, I was again expecting a much higher figure based on the AQUA CH5 anomaly, although the actual figure is still within the normal range.
    In the case of RSS, the NH anomaly is down to 0.381c, from a revised 0.432c (previously 0.449c) and the SH anomaly is up to 0.19c, from a revised 0.134c (previously 0.126c).
    Based on the above, I am still expecting very little change in the HadCRUT3 anomaly this month.
    What will be interesting will be to see how October continues. At the moment, the AQUA CH5 anomaly is pointing to a UAH of between 0.1c and 0.2c and a HadCRUT3 of about 0.4c, although obviously there is a long way to go.

  • Comment number 67.

    Spanglerboy @ #65

    "temperature is an intensive property and as such cannot be averaged."

    With respect, there is no rule saying you can't take an average of temperatures across a predefined area (or even the whole globe). The key point is that if mean temperature rises, it is indicative of an increase in the amount of heat energy......... and, of course, it is mean temperature ANOMALY which is used specifically because it is the change in temperature that is important. Also, as you may know, temperature anomaly tends to correlate well over quite large distances whereas absolute temperature does not.

    As you may be aware, Roger Pielke Snr has recently been arguing that ocean heat content is the only reliable metric for measuring warming. However, the problem with this is that only the top 700 metres of the oceans is properly monitored and even this record does not go back that far. The issue is discussed in this recent article at Realclimate:


    You'll note that Gavin Schmidt makes some pertinent remarks regarding why the surface temperature record is the one most heavily relied upon:

    "........ the surface temperature records are the longest climate records we have from direct measurements and have been independently replicated by multiple independent groups. I’m not aware of anyone who has ever thought that surface temperatures tell us everything there is to know about climate change, but nonetheless in practical terms global warming has for years been defined as the rise in this metric."


  • Comment number 68.

    62. Spanglerboy:

    SST usually refers to the top 1mm temperature of sea surface in most cases.

    It's a very unreliable guide to rely on as a measure of atmospheric warming/cooling.

    If you are referring to the 'earth system', then does this include such measures as minimum Arctic sea ice extent?

  • Comment number 69.

    Actually, HadSST2 seems to tie in very closely with HadCRUT3.
    The HadCRUT3 anomalies tend to be higher than HadSST2
    but over the last 30 years the trend in HadSST2 seems to slightly
    higher than HadCRUT3.
    I have found that I can estimate monthly HadCRUT3 from HadSST2 to within less than +/- 0.05c.

  • Comment number 70.

    69. QuaesoVeritas:

    I don't dispute that QV. I was replying to Spanglerboy @62. The point I was trying to make (and I see I used a poor choice of words) is that SST and surface air temperatures, which are closely related, are not the only means of gauging the planet's total energy balance.

    I'm sure you know that ocean heat content is not the same thing as sea surface temperature. Due to the volume of the ocean it is able to absorb enormous amounts of heat energy throughout its layers without any significant increase in temperature (something like 0.1C in the deep ocean over the period of recent warming).

    Another point I was making was that if we're talking about the total energy balance of the 'earth system' then things such as glacier reductions and sea ice loss also have to be taken into account.

    A hiatus in rising SST and SAT, even one spanning a decade or more, does not mean that the planet has stopped accruing heat. The imbalance between incoming and outgoing energy at the top of the atmosphere suggests that the planet is continuing to gain heat energy.


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.