« Previous | Main | Next »

All eyes on June after an 'average' May

Paul Hudson | 16:58 UK time, Thursday, 31 May 2012

The very warm spell which has dominated our weather in recent days has come to an end, with rain affecting much of northern England today.

It's a wet end to a month which has seen a remarkable change from extreme cold to extreme warmth.

And it's a good example of how statistics can hide big swings in the weather.

Because statistically temperatures, sunshine and rainfall for May have turned out to be around normal.

But this masks a very cold first half of May, with some sharp frosts and heavy rain which brought an end to drought restrictions in parts of Yorkshire; and a very warm second half of May which saw temperatures of 27C (81F) across the region.

Weather Action's widely publicised forecast of the coldest, or near coldest May in 100 years in parts of the country, which appeared in The Express, has turned out in the end to be wide of the mark - despite what was a very cold first half - because of the heat which developed in the last 10 days.

The switch from one extreme to another in May follows March which was one of the warmest, driest and sunniest on record, and April, which was the wettest on record, continuing the very polarised weather we have been experiencing.

All eyes are now on June - and there's a strong indication that the first half of the month will be very unsettled, with low pressure bringing showers or longer spells of rain, some of which may be heavy at times.

More specifically for the Jubilee weekend northern Britain should have mostly dry weather, but parts of southern Britain could experience a washout on Sunday, with some of the rain continuing into Bank Holiday Monday.

Just how far north the rain gets on Sunday is very uncertain at the moment, but there is a risk of it pushing as far north as the Humber before it returns south later in the day - although the precise northern extent is open to considerable doubt at the moment.

Follow me on twitter @Hudsonweather

Comments

Page 1 of 2

  • Comment number 1.

    Yes, indeed - how statistics can mask true variations and patterns!

    Once again - "Weather Action" speaks for itself. The most remarkable feature of their predictions is that people continue to believe them.

    So, June seems likely to start cool and changable - but how it finishes may be more significant. By old climatological rule of thumb -the general pattern for summer is often established towards the end of the month - the start of what is sometimes called "the European monsoon" when a more persistant westerly flow tends to set in after the ups and downs of spring - the position of the jet stream being of course crucial as to whether fine or very unsettled weather predominates.

    And yes Paul, yet another extreme month has followed others.

  • Comment number 2.

    I noticed tonight that there was a lot of cloud seeding going on in the evening skies. Does this activity make any difference to weather forecasts?

  • Comment number 3.

    We'll need to wait another few days for the official CET figure, but NetWeather run an approximation using nearby weather stations. According to them, the final May figure rounds to 12.5C, which is 1.2C above their 1971-2000 anomaly reference period.

    I notice Net Weather's figure is often higher than the official CET one. Even so, it would seem that in CET at least, far from being "the coldest in 100 years", average May 2012 temperatures were *above* the long term average, and possibly in the top 20 warmest of the past 100 years.

    I also take the point that in the first half of the month Weather Actions 'top 5 coldest' forecast at least was looking like a reasonable bet. But the bookies don't pay out at half time, and Piers Corbyn did specifically say "May", not "the first half of May".

    It'll be interesting to see the reaction - if any!

  • Comment number 4.

    Piers Corbyn, the man the Met Office fear. You can tell how much the alarmists worry about his accurate solar-based forecasts by how often they go on attack mode in this blog.

    He does this all without the hundreds of millions the Met Office takes from the taxpayer for its unnecessary GIGO supercomputers.

  • Comment number 5.

    Has anyone got a link or a copy of the actual forecast that he made? I'd like to see it.

    From reading comments here over the last few weeks, it seems to me that the cold was predicted reasonably well although it didn't last as long as thought. Were other forecasters also predicting a cold spell in May at the same time. That is, how far out on a limb was this forecast when it was made?

  • Comment number 6.

    @ Pingosan

    Is that the same Piers Corybyn who said on the 23rd that the warm weather would only last 2 days? Less attack mode, more LOL mode.

  • Comment number 7.

    If you make 'extreme' predictions (Coldest in 100 yrs, hottest year ever, -20 degrees last winter, etc) and it doesn't actually happen, then you should expect some comeback. You were happy for the publicity at your prediction and you have to take it on the chin when the world doesn't match up.

    Works for both sides.

  • Comment number 8.

    jkiller said . . "Yes, indeed - how statistics can mask true variations and patterns!"

    Which is why conclusions drawn from paleo temp reconstructions should be viewed with huge uncertainty. Or put another way, why claiming strong evidence of faster, stronger warming or highest global temps in 2million years (or whatever) is such a bag of . . .

  • Comment number 9.

    #3 newdwr54

    You may be correct for the final May CET but it would surprise me if it was 12C or over!
    http://www.metoffice.gov.uk/hadobs/hadcet/cet_info_mean.html
    Currently sitting at 12C up to the 30th so one more max and two min temps to take into account.

  • Comment number 10.

    and the MO/BBC prediction for May on the 17th April was? Piers Corbyn's prediction went awry on the 20/21st. No he did not correctly foresee the hot weather, but did anyone else in mid April?

  • Comment number 11.

    I have run a comparison between the daily CET temperatures of May 2012 and other May daily temperatures for the entire series (starting in 1772), and the correlation coefficients for months which come closest to this May are:
    1841 = 0.736
    1892 = 0.802
    1920 = 0.837
    All of the above months started cold and ended warm, with 1892 starting with a figure of 7.2c and ending on 18.8c, compared to 11.6c and 16c* in the case of 2012.
    * The figures for 2012 are still provisional and the figure for the 31st, which is not yet published, is estimated at 16c.
    The final monthly average figures for these months were:
    1841 = 12.7c
    1892 = 11.6c
    1920 = 11.8c
    2012 = 12.0c (provisional)
    So in terms of the historical CET record, the daily temperature variation this May is not unprecedented.

  • Comment number 12.

    #4. - PingoSan wrote:
    "Piers Corbyn, the man the Met Office fear. You can tell how much the alarmists worry about his accurate solar-based forecasts by how often they go on attack mode in this blog."

    Nobody could call me alarmist, but I don't think it is "attacking" Corbyn, to point out that his forecast was wrong. Is there any statistical evidence to support the assertion that his forecasts are accurate?
    That said, I believe that the M.O. also make unsubstantiated claims regarding the accuracy of their forecasts.

  • Comment number 13.

    #9. - ukpahonta wrote:
    "Currently sitting at 12C up to the 30th so one more max and two min temps to take into account."
    Actually, quite a lot more min. temps.
    The provisional figures tend to be over-stated because they do not necessarily include the lowest temp. for each day.

  • Comment number 14.

    #10. - Boanta wrote:
    "and the MO/BBC prediction for May on the 17th April was? Piers Corbyn's prediction went awry on the 20/21st. No he did not correctly foresee the hot weather, but did anyone else in mid April?"
    The point is that P.C. made the claim that he could forecast the temp. for May, and that it would be one of the coldest on record. Whatever method he uses apparently failed to foresee the warmer temperatures at the end of May, so I don't think it is unreasonable to point that out.

  • Comment number 15.

    QV - am I the only one who hasn't actually seen/read this forecast? Was it clearly intended as a May monthly average forecast or was it more casually worded?
    My view is that if no other forecasters saw the cold period coming then he deserves some credit.
    I think most of us regard long range forecasting as virtually impossible in detail due to the chaos in the system, but that doesn't preclude some skill being possible for more general outlooks. If Corbyn thinks otherwise, then maybe he's fallen for his own salesmanship.
    As you point out, the difficulty is measuring this skill. For example, I reckon that over the last 6 months my Hollybush has outperformed just about everyone including the met office!

  • Comment number 16.

    14: he has indeed recognised the failure to predict the week of hot weather and has amended his technique in that recognition. Most of the month was indeed cool and at times cold.

  • Comment number 17.

    15: He released the forecast on the 16/17th April. His predictions were for an unusually cold May in north and east Britain( not just England and not for western Britain). The implication here was for cold north easterly winds. For much of the month this was correct with several places in Scotland having record or near record minima. Even during the hot spell there was am easterly component in the winds or a col. In the last few days of May the 'pattern' returned to that which he predicted. You can always try the Weather Action site and check the comments section. There's even a phone number and you can always ask him yourself.

  • Comment number 18.

    Boanta - if your description of his forecast is accurate, then I would say his forecast had real merit. In six months time, people will remember April as generally wet and May as generally cold. I doubt that he would appreciate a call from me.

  • Comment number 19.

    "The coldest or near coldest May for 100 years in Central and East parts with a record run of bitter Northerly winds. Snow at times especially on high ground in NE / East. Spring put in reverse.

    Confidence of E / SE England mean temps: Coldest in 100yrs 80%; In 5 coldest in 100yrs 90%"

    No mention of the North in that prediction...

  • Comment number 20.

    I know Piers methodology is a well kept secret, but does anyone else here wonder if he isn't drawing from the work of Theodor Landscheidt? It might explain why his forecasts are occasionally close, if not quite, on the button - and lets be honest, who else's are? I do not want to trigger a slanging match on degrees of accuracy of the various parties - but it would be interesting to hear what others think of Landscheidts work and if it might be being used. If Piers is around, maybe he could comment too...

    newdwr54
    In case you haven't seen it, I have posted a further response on the previous thread.

  • Comment number 21.

    19: I believe NE stands for NORTH -east.

  • Comment number 22.

    19: The Cairngorm Ski Company reported excellent sking conditions and were very busy during May. Lying in North-east Scotland and Britain this would not suprise Piers Corbyn.

  • Comment number 23.

    #15. - lateintheday wrote:
    "QV - am I the only one who hasn't actually seen/read this forecast? Was it clearly intended as a May monthly average forecast or was it more casually worded?
    My view is that if no other forecasters saw the cold period coming then he deserves some credit."
    I am sure that someone here posted a link to the Weather Action website, but I can't for the life of me find it now.
    I am not sure if the quotes that others have posted are the entire forecast.
    From what I recall, some GFS forecasts (e.g. the NCEP one on the Ryan Maue site), were also forecasting a colder than average start to May, but I am not sure of the timing and the precise numbers.

  • Comment number 24.

    #16. - Boanta wrote:
    "14: he has indeed recognised the failure to predict the week of hot weather and has amended his technique in that recognition. Most of the month was indeed cool and at times cold."
    That's nice to know, but did he provide any details?

  • Comment number 25.

    To allow Piers Corbyn a right of reply he made a comment here including the admission of the unexpected warmth of late May:-

    http://climaterealists.com/index.php?id=9720&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+ClimaterealistsNewsBlog+%28ClimateRealists+News+Blog%29

  • Comment number 26.

    25: a very useful contribution and I hope this will help 24. Googling Weather Action also works and once on it's easy just to post a comment and ask.

  • Comment number 27.

    @21 Boanta

    Sorry I meant no mention of the North in his main 100yrs prediction. The bit with levels of high confidence, 90% chance of it being within the 5 coldest May's in 100 yrs. That prediction was for East/South East of England.

  • Comment number 28.

    5. lateintheday wrote:

    "Has anyone got a link or a copy of the actual forecast that he made? I'd like to see it."

    Hope this link works: http://www.weatheraction.com/displayarticle.asp?a=450&c=5

  • Comment number 29.

    We holidayed in Majorca mid April and it was overcast cold and wet. About the same time last year the sky was cloudless and it was warm. This year on a cold murky day on the islands east coast we had lunch and the conversation with the waiter turned to the weather. I explained that the ‘extreme’ weather was being caused by excessive jet stream waviness. I think he got the idea?

    Then coincidently an article from Yale University appears on the EcoGeek site

    http://e360.yale.edu/feature/linking_weird_weather_to_rapid_warming_of_the_arctic/2501/

    Said much the same thing only it added

    “Theory tells us that a decrease in the west-east flow tends to slow the eastward progression of waves in the jet stream. Because these waves control the formation and movement of storms, slower wave progression means that weather conditions will be more persistent. In other words, they will seem more “stuck.””

    Paul you’re not explaining yourself too well. The met office really needs to show the position of the jet stream more often but you can infer its location from the pressure charts. Long periods of hot and cold weather are going to be the norm; so much for a temperate climate.

  • Comment number 30.

    Having read Corbyn's comments in todays Guardian
    http://www.guardian.co.uk/environment/blog/2012/jun/01/whatever-happened-coldest-may-years
    I think Corbyn should be hired by the England football team - his skill with goalposts would be invaluable!!

  • Comment number 31.

    Piers is sometimes right, sometimes wrong, but he's normally contrite when wrong...... given the lack of millions of pounds worth of super computer I'd be inclined to cut him a little slack.... regardless, the best way of telling what the weather in short term is still going to be looking out the window and a good old fashioned barometer ;-)

    Long/medium term, you can take your pick, the Met Office generally tend to a warmist persuasion..... whilst Piers tends to plough his own rather secretive furrow.....

    He gets a lot of stick, but he's definitely no worse than our rather over assertive civil servants.

  • Comment number 32.

    #25. - jazznick wrote:
    "To allow Piers Corbyn a right of reply he made a comment here including the admission of the unexpected warmth of late May:-
    http://climaterealists.com/index.php?id=9720&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+ClimaterealistsNewsBlog+%28ClimateRealists+News+Blog%29"

    I am afraid that I cannot get that link to work.

  • Comment number 33.

    Met. Office anomaly maps show that mean temperatures for most of the North and East of G.B. were within +/- 0.05c of the 1961-1990 average:
    http://www.metoffice.gov.uk/climate/uk/anomacts/
    The remainder of the country was between + 0.05c and +1.5c
    Piers Corbyn seems to have been generally correct for the month as a whole, but it the general accuracy has been overshadowed by extreme talk of "coldest May for 100 years".

  • Comment number 34.

    Please note that the above map seems to have defaulted to "Mean daily maximum temperature" and the 1971-2000 anomaly.
    To see the "Mean temperature" and 1961-1990, you will have to select those.

  • Comment number 35.

    Confirmed by the MO coldest May ever, Pity you can't show screen shots:
    http://www.metoffice.gov.uk/hadobs/hadcet/cet_info_mean.html

    May -327.7C Anomoly -338.8C

    I'm sure that it will be corrected soon, at least I hope so.

  • Comment number 36.

    35. ukpahonta:

    Looks like I have to take a few things back!?

  • Comment number 37.

    33. QuaesoVeritas:

    Your link is actually to 'mean daily *maximum* temperature', and the reference period is 1971-2000.

    If you select 'mean temperature' from the climate variable list you'll see that almost all of South Central England was +0.5 to +1.5C above the anomaly reference for May 2012. Using the WMO recommended 1961-1990 anomaly the area of above average temperatures is even broader.

    Corbyn's prediction was for "The coldest or near coldest May for 100 years in Central and East parts with a record run of bitter Northerly winds."

    Therefore I think mean temperatures are called for, as is the longer term anomaly reference period.

  • Comment number 38.

    34. QuaesoVeritas:

    Re my 37: sorry, I missed your 34 before posting 37.

    Would you agree with my assessment that using the mean temperature and 1961-1990 makes Corbyn's forecast for May 2012 more or less completely wrong?

  • Comment number 39.

    #37. - newdwr54 wrote:
    "Your link is actually to 'mean daily *maximum* temperature', and the reference period is 1971-2000. "
    Please see my subsequent post.
    For some reason, the map seems to default to mean maximum and 1971-2000.

  • Comment number 40.

    Sorry, I missed your #38 before my #39 !
    Now I have to wait 2 min before I can post this!

  • Comment number 41.

    #38. - newdwr54 wrote:
    "Would you agree with my assessment that using the mean temperature and 1961-1990 makes Corbyn's forecast for May 2012 more or less completely wrong?"
    As I said in my post #33, he was *generally* correct in terms of the overall weather pattern, but completely wrong in terms of the very cold temp.
    I think this is the case whether you use the 61-90 or 71-00 averages.
    Actually, a more meaningful comparison would probably be vs 1911-1940.

  • Comment number 42.

    41. QuaesoVeritas:

    If you select 'Mean daily minimum temp' using either 61-90 or 71-00 you should see that the entire SE of England was above average, and in places +1.5 to +2.5 above average.

    There is no noticeable cold bias in the East using this option, i.e. cold days in the east weren't particularly colder than they were elsewhere. Only in the extreme north of Scotland were mean daily minimum temps below average in May.

    The best we can say is that warm days in May weren't *as* warm in parts of the SE as they were elsewhere - but cold days weren't any colder than elsewhere either.

    No doubt Weather Action will count this as a sweeping victory.

  • Comment number 43.

    If he had said 'May will have periods of pretty cold weather' then he would of been right. He didn't, instead he went for a headline grabbing statement which didn't come true.

  • Comment number 44.

    Oh and the reason the forecast fell over is because we are entering a mini-ice age... no really!

  • Comment number 45.

  • Comment number 46.

    I would like to see some global warming. Up to 2000 and just past it we were seeing some warming, but that was due to the sun going through a very high peak. Since 2006 there hasn't been a decent summer. It is all a big myth, by governments to tax us and business to use it as a way of making money out of us.

  • Comment number 47.

    46. Tim wrote:

    "Up to 2000 and just past it we were seeing some warming, but that was due to the sun going through a very high peak."

    But since 2000 solar activity has dropped and temperatures have risen.

    http://www.woodfortrees.org/plot/uah/from:2000/plot/sidc-ssn/from:2000/normalise/plot/uah/from:2000/trend/plot/sidc-ssn/from:2000/normalise/trend

    If it's just the Sun, then how can this be? If the Sun is the cause of the warming, then has it warmed during the period in which since solar activity fell?

  • Comment number 48.

    #46. - Tim wrote:
    "Up to 2000 and just past it we were seeing some warming, but that was due to the sun going through a very high peak."
    I'm not sure exactly what you mean by that.
    Do you mean a peak in sunspots?
    The highest peak in the 12 month m.a. sunspot number was in 1958 (cycle 19), followed by a much lower max. in 1969 (20). The peaks around 1980 (21) and 1990 (22), were relatively high, but not as high as in 1958. The latest peak, around 2001 (23), was lower than those around 58, 80 and 90, and not much higher than 69.
    Also, the 10 and 20 year m.a. ssn both peaked around 1965, so why did it take until 1998 for temperatures to peak, if that was related to ssn?

  • Comment number 49.

    After being relatively low until about the 28th., the AMSU CH5 temperature anomaly has taken an upward turn over the last 3 days, but this is too late to make a great deal of difference to the overall May anomaly.
    Based on the figures up to the 31st, I am estimating the May temperature anomaly figures to be as follows:
    UAH = 0.252c +/- 0.04c
    RSS = 0.225c +/- 0.03c
    HADCRUT3 = 0.40c +/- 0.05c
    The above are all slightly lower than the respective anomalies for April, but the upward turn in temperatures during the last few days continues, June may see a return to higher anomalies.

  • Comment number 50.

    Not to labour the point too much further, but according to MO data, the regional minimum daily average temperature recorded in 'England SE/ Central S' for May 2012 was +8.2C.

    This places it inside the top ten *warmest* May minimum daily average temperatures recorded in past 100 years.

    http://www.metoffice.gov.uk/climate/uk/datasets/#

  • Comment number 51.

    #50. - newdwr54 wrote:
    "Not to labour the point too much further, but according to MO data, the regional minimum daily average temperature recorded in 'England SE/ Central S' for May 2012 was +8.2C."
    Personally, I think that in terms of the Piers Corbyn forecast, the mean temperature for E/NE England is probably more relevant, but even that puts it in the top half of the temperature range since 1910.

  • Comment number 52.

    My only interest in short term weather is how it effects the growing season, (listen to the shipping forecast and work it out) however the following may be of interest to others:-

    “ Who are you calling a charlatan?”

    http://www.bishop-hill.net/blog/2012/6/3/who-are-you-calling-a-charlatan.html

    “For Monbiot to call Corbyn - a private citizen - a charlatan while keeping silent on publicly funded scientists seems...opportunistic.”

    I do hope George is still tending his raised beds in deepest Welsh Wales.

  • Comment number 53.

    Remember as we float through the galaxy, the density of the matter changes and can reduce or allow more sun rays to hit us, on top of the sun going through warming and cooling periods. The best evidence is the temperature of the earth in 1000A which was warmer than now, the lack of sun in the dark ages when the Thames froze over. Most evidence is False and a diversion.

  • Comment number 54.

    To illustrate how different things have become the timing of the disappearance
    of the 528 line is striking. The 528 thickness line shows where the temperature
    between 1000 and 500 mb has an average temperature of 0 degrees c. I definitely remember that on the Met Office charts from say Newfoundland to Western Russia a pool of 528 (i.e. deep cold air) always clung on till the end of June and reappeared about Mid August.
    Last year was very different the 528 vanished on the 8th June. and didn't reappear till the last week of August (it vanished again during that extraordinary late Sep/early Oct heatwave). This year it disappears again just as early. Looks to me very suspiciously like a step change in the way the atmosphere is behaving -perhaps hand in hand with the low Arctic ice extents we are also getting?

  • Comment number 55.

    Apparently 46 people, including 6 of those taking part in the Diamond Jubilee river pageant had to be given treatment by medics after suffering from the effects of cold and wet weather.
    http://www.itv.com/news/update/2012-06-03/pageant-crews-and-spectators-taken-to-hospital-after-downpours/
    Having said that, it may have been partially due to them not wearing appropriate clothing for the weather.
    No doubt the cold/wet weather will be blamed by some on "climate change", or "global weirding", but if it had been exceptionally warm, and the spectators had been treated for heat-stroke instead, it would have been blamed on the same thing. Actually, the weather was very similar on Coronation Day, in 1953, but more people wore raincoats and hats in those days.
    http://www.weatheronline.co.uk/reports/philip-eden/Coronation-Weather.htm
    Those of us who are old enough, will know that the British weather has always been this variable, and it no doubt always will be.

  • Comment number 56.

    The May UAH anomaly is 0.289c, slightly lower than the April figure of 0.299c.
    While this is just within my estimated range of 0.252 +/- 0.04c, I must admit that I am surprised that it is so high.
    No doubt when the daily UAH figures are published, and they can be compared with the daily AQUA ch5 figures, that will shed some light on the situation.
    The N.H. anomaly is up from 0.413c to 0.439c and the S.H. is down from 0.185c to 0.139c.

  • Comment number 57.

    Has Dr Roy gone back to a 4th order polynomial fit? I wonder why?

  • Comment number 58.

    A large Coronal Hole has appeared on the Sun and the effects are due on Earth 5-7th June. So if the solarists are correct we might expect noticeable differences from standard weather predictions. The Countryfile forecast indicated a cold air mass over us, and indeed most of the North Atlantic for the whole week, with the Jet Stream well to the south. So will be easy to see any dramatic flipping.

  • Comment number 59.

    #57. - john_cogger wrote:
    "Has Dr Roy gone back to a 4th order polynomial fit? I wonder why?"
    Back to?
    When did he last use that?

  • Comment number 60.

    #58. - Boanta wrote:
    "A large Coronal Hole has appeared on the Sun and the effects are due on Earth 5-7th June."
    Apparentlly it looks a bit like a chicken!
    If this appears on the news, everyone will be running around like headless ones!
    http://www.spaceweather.com/

  • Comment number 61.

    @59 QuaesoVeritas

    newdwr54 talked about it in the last thread. 2009 was mentioned for the 4th order, then no trend line for a while, then 3rd order up to last month.

  • Comment number 62.

    57. No, it's just the way he walks.

  • Comment number 63.

    56. QuaesoVeritas wrote:

    "The May UAH anomaly is 0.289c, slightly lower than the April figure of 0.299c.
    While this is just within my estimated range of 0.252 +/- 0.04c, I must admit that I am surprised that it is so high."

    I was just about to congratulate you on your accuracy! If Piers Corbyn had predicted that one as accurately as you just did he'd be demanding a Knighthood.

    One point of interest re the UAH data is that Anthony Watts, last time I checked 5 mins ago, has still not updated the UAH chart in his 'Global Temperature Page' (Reference Pages) with April's UAH value, never mind May's. Maybe he was holding out for a drop in May?

  • Comment number 64.

    61. john_cogger wrote:

    "2009 was mentioned for the 4th order, then no trend line for a while, then 3rd order up to last month."

    That's right John, and well spotted.

    Actually I think it's a 'typo' by Spencer. The published chart looks a lot more like he's added a 3rd order polynomial trend line, not a 4th. A 4th order one makes the line level out and not appear to dip - and we can't have that, can we?

    QV and others, if you want a bit of fun, try adding a 6th order polynomial trend line to the UAH data. This was the trend line of choice for 'sceptics' such as Allan MacRae during late 2008, when there was a temporary cooling period globally. (See for instance: http://icecap.us/images/uploads/uah7908.JPG )

    You'll see why it has now been quietly dropped by 'sceptics'.

  • Comment number 65.

    #63. - newdwr54 wrote:
    "I was just about to congratulate you on your accuracy! If Piers Corbyn had predicted that one as accurately as you just did he'd be demanding a Knighthood."

    To be fair, it's an estimate, not a prediction, since I have the advantage of knowing the AQUA CH5 figures, but while it should be neutral, the actual figure always seems to come out on the high side.

  • Comment number 66.

    For interest/balance perhaps this item from Climate Realists will explain why Piers'
    forecast went a bit awry at the end of May.

    http://climaterealists.com/index.php?id=9746

    The Sun is the major driver of Piers' forecasts so the GLE on 17th May was pivotal.

  • Comment number 67.

    @66 jazznick

    Piers only has an accuracy of 80% from that article (with a now standard dig at the Met Office) yet he was 90% sure that May would be within the 5 coldest. Is that 90% of 80%?

  • Comment number 68.

    @67 john_cogger

    Piers has every right to dig at the MO. To claim to forecast anything while ignoring solar influence is just expensive guesswork based on the movement of pressure systems and the speed/direction of the jet-stream alone.

    You are directing your comments at a scientist who is trying to work out the link between what the sun/moon does and HOW this pushes the jet-stream around, to put it at it's simplest. He combines current solar influences with historic weather data and what the sun/moon effect was at the time in order to search for patterns. The more he does this, the better he gets.
    The 22yr magnetic cycle seems to be key here.

    His aim is to predict serious climate switches where there is a potential danger to crops, property and life itself, not to tell you if it's safe to put your washing out tomorrow.

    Piers' recent forecast fell down because the sun threw a rare GLE at us. Once this passed we went back to the cold, wet, hail again.
    Give the guy some credit for not just accepting that 'weather just happens' in a totally chaotic manner and that there are solar effects that can warn us of dangerous/unusual weather to come.

    To it's credit the MO are belatedly intending to set up a Space Weather section so they clearly see some benefit in looking outside our atmosphere for climate drivers.

  • Comment number 69.

    Yes May was 'average' as described by the UKMet office. Our temperatures at the moment are controlled by the jet stream which is south of us so keeping the available heat south of them. We are getting the cooler air masses from further north.

    Will these jet streams go north? Good question and probably based on solar output in some way. The Sun is doing its pole change around now so magnetic output is down and it is the magnetic part of solar output that is important.

    So wait and see. Are the solar scientists right or the AGW activists? I bet on the science.

  • Comment number 70.

    @68 jazznick

    "Piers' recent forecast fell down because the sun threw a rare GLE at us. Once this passed we went back to the cold, wet, hail again."

    So how did his drought prediction for April fall down? The cold November? The gales in January? The white Christmas? And those are just misses on the UK forecast.

    Or a prediction/statement of global cooling since 2003?

    These rare events keep screwing up the forecasts on a monthly basis.

    And he isn't a scientist until he starts publishing data or papers. Up to the point he does that he is a businessman with a self interest to slam the MO.

  • Comment number 71.

    Also the GLE hit earth on the 17th. The 'heat wave' was from the 23rd and lasted over a week. Boy that took a long time to get past.

  • Comment number 72.

    Oh also remember is was Piers who said that the heat wave would only last 2 days. He forecast that on the 25th, 2 days into his 2 day heat wave.

  • Comment number 73.

    @69 John Marshall

    Thank you John - just the comment I would have hoped for from the more open minded amongst those here.

    Clearly others @ 70+ think the MO can do no wrong, perhaps because of their stance on 'other matters'.
    However the MO establishment have nothing to be proud of:-

    http://thegwpf.org/the-observatory/4468-met-office-climate-forecasts-always-wrong-but-never-in-doubt.html

    http://autonomousmind.wordpress.com/2012/05/06/met-office-forecasting-produces-another-epic-failure/

    Something is clearly wrong with the way weather/climate is currently being predicted and to expect more computers and bonuses to solve it is not going to work. Eventually solar influences are going to feature in weather and climate forecasting (hence the MO Solar Weather plans) and the sooner this is realized the better for the planet and it's people regardless of their views on CO2.

  • Comment number 74.

    @73 jazznick

    You keep mentioning the MO failures? Not sure how relevant they are to Piers failures? Then again he thinks they are 'lying scum' so maybe there is a link?

    At no point have I said the MO are perfect or used them as a comparison against Piers. Yet you keep bringing them up? Maybe it's because of their stance on other issues?

  • Comment number 75.

    @74 john_cogger

    If you actually read my final line above you will see that CO2 can be left out of the picture when it comes to this subject; regardless of Piers' colourful language and views on 'that matter'.

    What I am trying to encourage is the urgent exploration of climate influences beyond the conventional. The status quo will not do. Building ever larger ivory towers of computers will only lead to wrong answers being arrived at more quickly. There is a big gap in our understanding of the area that Piers works in and for his work to be dismissed by you is short-sighted.

    Clearly there is something significant going on here. It's about time it was seriously researched by the MO/Hadley Centre rather than our money being wasted on flogging the same old models.

  • Comment number 76.

    75. jazznick wrote:

    "There is a big gap in our understanding of the area that Piers works in and for his work to be dismissed by you is short-sighted."

    If he would open up his methods to peer reviewed scrutiny maybe it would help his cause?

    As for the link you presented earlier, which blamed the failed forecast on "an M-class flare" on the Sun: If an apparently random event like this on the Sun can throw off a specific prediction so easily, then forecasters have no business making 90% confidence predictions.

    In other words, if these "M-class flares", etc can randomly occur, and if they really do have a major impact on global atmospheric conditions, then all long-term weather forecasts are *useless*. They can't be made with *any* stated confidence level, because they rely in part on totally unpredictable events.

    "M-class flares" and the like make Piers and co redundant.

    (I seem to recall that a similar 'unforeseen' solar event was blamed for Piers' hopelessly wrong UK and US winter forecast for 2011/12.)

  • Comment number 77.

  • Comment number 78.

    77. ukpahonta:

    Is there some suggestion that ENSO has been ignored by the IPCC or mainstream climate science?

  • Comment number 79.

    #78 newdwr54

    There is some suggestion that there is a connection between ENSO and Aqua Channel 5 Temperature with a three month lag. I thought that interesting!

  • Comment number 80.

    #77. - ukpahonta wrote:
    "This should be right up your alley:"
    I have only had time to have a quick look at both links, but I am going to have to spend more time on the second one.
    Obviously there appears to be be a correlation between the ONI and the AQUA CH5 anomaly but I honestly can't understand what it means when it says that "ONI leads temperature by 3 months". I don't think that it is as clear cut as that.
    Sometimes the temperature appears to change before ONI, for example in the latest increase in temperatures.
    If I can, I will try to replicate the numbers myself, as I can't tell precisely from the graph, where changes in the temperatures actually occur.
    I am not entirely sure if an r-squared value of 0.37 is significant, but even if it is, I don't think it necessarily demonstrates causality, one way or another.

  • Comment number 81.

    My instinct is that you would get a similar r-squared figure, whether you used a lag of +3 months, -3 months, or zero.
    I might try that, when I get time.

  • Comment number 82.

    maybe piers corbyn was right, cet page currently shows it was -327.7C on average in May
    http://www.metoffice.gov.uk/hadobs/hadcet/cet_info_mean.html

  • Comment number 83.

    #82. - quake wrote:
    "maybe piers corbyn was right, cet page currently shows it was -327.7C on average in May"
    I think that ukpahonta pointed that out a few days ago.
    It seems to have been fixed now.
    Probably delayed due to the long weekend!

  • Comment number 84.

    re my posts #80 &81,
    Needless to say, my instinct was wrong!
    I managed to more or less replicate the AQUA CH5/ENSO correlation data and graphs on the lukewarmplanet blog and get similar results. Some of the monthly ONI figures look slightly different to those on the graph but the data may have been changed since the lukwarmplanet blog was written.
    I understand now by the ONI data leading the CH5 data by 3 months, but actually it looks to me more like it is 6 months on the graph. That said, I get an figure for
    r-squared of 0.362 for a 3 month offset (at June 2011). The difference between that and the 0.37 quoted on the site may be due to changes in the data.
    I actually calculated the r and r-squared values for a range of offsets between ONI and AQUA CH5, of +6 to -6, where +3 corresponds to the 3 month offset used in the graph. The results were:
    OFFSET r r^2
    +6 0.572 0.327
    +5 0.611 0.373
    +4 0.627 0.393
    +3 0.601 0.362
    +2 0.528 0.279
    +1 0.438 0.192
    0 0.314 0.098
    -1 0.173 0.030
    -2 0.053 0.003
    -3 -0.059 0.004
    -4 -0.143 0.020
    -5 -0.206 0.042
    -6 -0.237 0.056
    Sorry for the formatting.
    It can be seen from the above that the strongest correlation is actually for an offset of 4 months and that the correlation falls towards an offset of zero, and is actually negative for an offset of -3 and beyond.
    If the above figures are plotted, they actually produce a couple of nice curves, which suggest that the optimum offset is possibly a little over +4 months.
    Of course, this opens up the prospect of using ONI to approximately forecast the likely monthly AQUA CH5 temperature, and also the UAH and other anomalies, for about 4 months ahead.
    That said, on further reflection, perhaps an r-squared figure of around 0.39 is as much as you can expect, given the chaotic nature of the climate system, and I wouldn't expect such forecasts to be particularly accurate, but I might give it a go!

  • Comment number 85.

    I have just noticed that the ONI index covers 3 month overlapping periods, e.g. DJF, JFM, FMA etc, so this may account for the apparent 6 month offset on the graph on the lukewarmplanet blog, and the fact that I get a higher correlation at an offset of +4 than +3 months, but I haven't worked out exactly what the full implications are yet.

  • Comment number 86.

    There are also two versions of the files, due to the changes in the method of calculation of the index.
    Urgh!

  • Comment number 87.

    QV,

    Re your above recent posts; the relationship between AQUA CH5 and ENSO is something I've been collating for a few months.

    I used correlation analysis and found that 11-12 weeks gave the best fit. I've been able to semi-successfully forecast week-ahead temps in CH5 using this method for a few weeks now. It's not perfect, but it's not too bad either - to date.

    Something to bear in mind is that ENSO just amplifies CH5 temps - there are also the normal annual temperature fluctuations to consider.

    However, based on this relationship, my crude forecast is that temps in CH5 should on average increase at an above normal rate for about two weeks (up to the week ending 17th June), then return to more or less normal for the time of year for a week or two.

    I realise that, Like Piers Corbyn, I'm putting my neck on the block here. However, unlike Piers, I promise to admit if my forecast turns out to be drivel. And I don't care really; I just do it for fun, not for a living.

  • Comment number 88.

    #87. - newdwr54 wrote:
    "Re your above recent posts; the relationship between AQUA CH5 and ENSO is something I've been collating for a few months."

    Thanks, I was aware that you were doing something along those lines, but I didn't know how formally you were doing it.
    Which ENSO measure are you using to calculate the correlation?
    There appear to be two versions of the ONI, the most recent of which appears to have been calculated using multiple 30 year base periods, in an attempt to remove any warming trend. I think that would make it less useful for predicting CH5, as it would tend to under-estimate recent temperatures, but I am going to do some comparison between both versions.
    I believe that the monthly nino-3.4 index also uses the new multiple base periods, which seems to have the effect of reducing the most recent anomalies.
    At the moment I am just dabbling, so you probably know more about this than me.

  • Comment number 89.

  • Comment number 90.

    This article in Nature Geoscience, seems to suggest that a significant part of any recent sea level rise may be due to "unsustainable groundwater use".
    If I interpret this correctly, it seems to mean that we are extracting groundwater at an unsustainable rate.
    http://www.nature.com/ngeo/journal/vaop/ncurrent/full/ngeo1476.html
    Of course, there will be some who argue that the ultimate cause is "climate change", but in reality I think it is mainly due to exponential growth in the human population and economic growth. Without that growth, any minor variations in climate would be manageable.

  • Comment number 91.

    rob@89 - looks like you've got no takers. The silence is deafening.

  • Comment number 92.

    #73 Thank you jazznick.

    I used to work using Met. Office information, I flew in the RAF, and found it wanting. They have not improved despite the computer power. They also refuse to believe observations if these fall outside their models. I was told once that an observation as to Cb height limitation was wrong and I must have had the height wrong because their models stated something else. They never learn.

  • Comment number 93.

    88. QuaesoVeritas:

    I'm using Australian Met Office NINO3.4 weekly: http://www.bom.gov.au/climate/enso/nino_3.4.txt

    It's available from here, usually on a Tuesday (make sure you select 3.4 from the drop down list and select 'data sorted by date': http://www.bom.gov.au/climate/enso/indices.shtml

  • Comment number 94.

    Corbyn's method has been proven in peer review literature:

    Early Weather Action (Solar Weather Technique) skill was independently verified in a peer-reviewed paper by Dr Dennis Wheeler, University of Sunderland, in the Journal of Atmospheric and Solar-Terrestrial Physics, Vol 63 (2001) p29-34.

    If you want to read summary of the report then look in the comments by me here:

    http://www.bishop-hill.net/blog/2012/6/3/who-are-you-calling-a-charlatan.html

  • Comment number 95.

    #94. - jackcowper wrote:
    "Corbyn's method has been proven in peer review literature:"
    Didn't that just relate to gales over a 2 year period?
    I don't consider Piers Corbyn a "charlatan", but I just require more substantive statistical evidence than others apparently do.
    I am also not uncritical of the UKMO, but if they had predicted one of the coldest May's in 100 years, I would be critical of them too, with good reason.

  • Comment number 96.

    #93. - newdwr54 wrote:
    "I'm using Australian Met Office NINO3.4 weekly: "
    Thanks, do you know if the weekly figures are also based on multiple base period?
    I have been looking at ONI in relation to predicting HadCRUT3.
    Actually, using last months HadCRUT3 figure ALONE to predict this months (with no adjustment), produces a long-term correlation of 0.93, but I have so far been unable to improve on that by any adjustment of the prediction using ONI.

  • Comment number 97.

    94. jackcowper:

    Wheeler's paper was quoted by Leo Hickman in the Guardian last week: http://www.guardian.co.uk/environment/blog/2012/jun/01/whatever-happened-coldest-may-years?newsfeed=true

    It studied Weather Action forecasts of *gales only* over a two year period in the 1990s. The study found a 52% success rate. Quoting directly from the Wheeler (2001), Hickman says:

    "Superficially these data suggest a degree of forecast success. This is, however, to overlook some sources of ambiguity. One of the most important of these is the question of seasonality of gales, which are more frequent in winter than in summer. Indeed the majority of successes in category d [(no forecast-no event] can be attributed to those summer periods when no gales were forecast — and none occurred. It is, self-evidently, not very difficult to predict successfully the non-occurrence of a rare event!"

    A peer reviewed assessment of Weather Actions winter UK/US 2011/12, and UK May 2012 forecasts might find rather less flattering results.

  • Comment number 98.

    depends on who does the peer review. . .

  • Comment number 99.

    Peer review in climate circles is usually done by one pals, just ask Mann, Jones, Briffa et all.

    It's a pity that Hickman cannot be as balanced in his reporting, but if you will read the Guardian - what do you expect?

  • Comment number 100.

    To John Cogger~70 - Exactly, Corbyn is a businessman, not a scientist - until as you say, he presents his data and theories for scientific scrutiny. "Trade secrets" are just that - a good way of making money.

    to newdwr54 ~ 76 - Precisely! If the sun is in the habit of throwing out "unpedictable" flares or whatever - how can it possibly be of any use for making reliable forecasts? As you say - this has been blamed before.

    But by all means, people, pay your money for something that is logically inconsistent and unverified. Personally I'll stick with Russell Grant.

 

Page 1 of 2

BBC iD

Sign in

BBC navigation

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.