Piers Corbyn, while seemingly a bit eccentric, has the distinction of being the only man to have this headline:
The man who repeatedly beats the Met Office at its own game
Armed only with a laptop, huge quantities of publicly available data and a first-class degree in astrophysics, he gets it right again and again.
Well, now he has competition from New Zealand with yet again another guy armed with a laptop and public data, as Jo Nova reports below:
Laptop beats Met Supercomputer: SOI index (at record high) scores a win.
Back on August 6, 2010, when the UK BOM was predicting a warm winter, and every Met Agency in the West was already declaring that 2010 would be the hottest year ever, Bryan Leyland predicted (on a global scale) that before the end of the year, there would be significant cooling. As you can see from the chart, this is exactly what happened.
The UK Met Office has a gigantic supercomputer, 1,500 staff and a £170m-a-year budget, but a retired engineer in New Zealand armed only with Excel and access to the internet and with the McLean is et al 2009 paper, was able to get it right.
Parking the SOI index (the blue line) 7 months into the future suggests things may get cooler still as the temperature (red line) often follows the trend. (Click for a larger image.) Note, the SOI is shifted 7 months forwards in time, and the scale is inverted.
Before anyone scoffs that the El Nino’s are usually followed by cooling, and the SOI indicator is well known, ponder that the well fed agencies of man-made-climate-fame weren’t telling the public that a big-chill was on the way and they ought to stock up on salt and red diesel. (And maybe take their own deicing fluid to the airport.*)
Read it all here
Another excellent article on Piers.
I wonder if I may point to an article of my own concerning the aggressive and bigoted Greenfyre blog?
http://theanti-politician.blogspot.com/2011/01/dousing-damp-greenfyre.html
Thanks!
Greg
orkneygal says:
January 12, 2011 at 3:21 am
Is this the same British weather office that General Eisenhauer relied upon to send the Soldiers of Democracy across the Channel into Normandy to begin the liberation of Europe?
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++
orkneygal, good one. Now, all we need is to reverse the process, send in the soldiers of democracy back to the UK, to the Met Office to be exact, and liberate us from the cli-myth shamans. They should be in prison right now, paying for their lies, damned lies and cooked weather predictions that caused thousands, millions of Britishers to suffer, some even die of the cold due to the unpreparedness of whoever should have been prepared for the biggest freeze-over ever recorded in UK history.
Robert Ellison says: “There are obvious and potentially serious risks from anthropogenic greenhouse gas emissions in a chaotic climate ” followed shortly by “rather than tediously repeat … nonsense”
You would do well to heed your own advice. There are no more “serious risks” from a warmer climate than from what is normal. In fact, I think the risks are much less. Instead, you claim the “serious risks” are “obvious”. How about showing proof instead of waving your arms and essentially screaming … the sky is falling.
Mike McMillan says:
I predict we have twenty more years of mostly cold left in this 30-year half cycle that began with the peak of 1998. Unless 1998 was the end of the Holocene (hedging my bets)
I’m more of the opinion the previous 30 year half cycle ended in 2005. The 1998 El Niño disguises the fact that the warming continued for a few more years, just as the 2010 El Niño disguises the current cooling trend that began 2006. That gives us even more years of cooler temperatures to look forward to.
[Another] Roy wrote:
“It really doesn’t matter that Piers Corbyn or Bryan Leyland once made a claim about the future and it later happened that way. What would matter is if they do it repeatedly and much more often than chance. ”
I must disagree with my namesake. If Piers Corbyn had said something along the lines of “I think it is going to be a rather chilly winter” that would not have been very impressive (even though it would have been more accurate than the Met Office’s “barbecue summer” forecast of a couple of years ago) but what he wrote was:
Winter Dec to Feb inclusive in Britain and Europe will be exceptionally cold and snowy – like hell frozen over at times – with much of England, Germany, Benelux and N France suffering one of the coldest winters for over 100 years.
Weather forecasting is not an exact science yet and it is possible that the rest of the winter might turn out very mild. However Corbyn was certainly right about December which was the coldest in Britain for a century.
What are the chances of getting that right by chance?
@Roy, Corbyn did not base his claims on an isolated flash in the pan as you imply. Plenty of evidence of previous skill can be found on the Weather Action site. If you’ve not seen any so far, you’ve just not looked properly – try this link: http://www.weatheraction.com/pages/pv.asp?p=wact5&fsize=0
Me thinks Hansen will adjust it….
“I’m as disappointed with the Met Office as anyone living in the UK”
At a 170 million pounds your share would be a pound per year? Yeah, no more than your support for a half-wit Prince of Wales, but a fraction of your contribution to wind-generated power, and a mere pittance in comparison to your support of the PIIGS and a failed EU.
We are humbled in witness to your wisdom.
I am normally just a reader and not a poster but as an Economist I can relate this article to similar predictive schemes in my field. By the end of the 1970s large scale complex economic forecasting models were the rage. However, a simple seven equation reduced form model developed by the St. Louis Federal Reserve bank to predict key policy variables consistently out performed the big models.
Complexity is not accuracy particularly when there are more unknowns than knowns and your equations are approximations to begin with. Most forecasts degenerate after a few time steps anyway as you go outside the sample data. When I ran an analysis shop I used to ask prospective analysts if they believed their model results. If they said yes they didn’t get the job. Models are merely tools to help us understand phenomena. The rarely give real answers.
December 2010 in the UK was not only the coldest month since Feb 1986 but had 39% more sunshine than average.
Meanwhile in Newfoundland….
Robert Ellison says:
January 12, 2011 at 2:00 am
There are obvious and potentially serious risks from anthropogenic greenhouse gas emissions in a chaotic climate –
CHAOS IT IS ALWAYS IN THE MIND OF THE BEHOLDER….”GOD DOESN’T PLAY DICE”
If we do not recognize general laws, it means simply WE do not know them, we are ignorant of them, or perhaps we have been deceived by a fanatical “creed”.
Lucy Skywalker says:
January 12, 2011 at 2:00 am
the science behind his work needs to become public
I have always appreciated you Lucy. You bring out things that show you have valid questions, and you are thinking. I like what you do, so, I hesitate to say this, but it seems unfair to Piers Corbyn that he should have to reveal how he does his work. It is as unfair as to have asked Colonel Sanders what his 11 herbs and spices were. I guess anyone who wants to know how he does it, like the Met, should start their own study and figure it out instead of expecting him to just tell. The Met has that super computer and a pretty building to work in. They should be able to do it.
The Met building:
http://wattsupwiththat.files.wordpress.com/2009/11/hadley-metoffice1.jpg
although I give credit to the forecaster, I must point out that anyone with a dartboard could outpredict the Met Office.
Sorry, but I can’t handle the term “SOI index”. That’s redundant. Reminds of when an actor in the part of a scientist says “RPMs”. Drives me nuts.
Let us assume that I wish to come up with a method of finding 7% of any number, which I will call N. If I divide N by 100, this will give me 1% of N. Now, if I multiply this 1% of N, which I’ve just found, by 7 this will give me 7% of N.
Having come up with this delicious solution, I decide to let my computer do the arithmetic. I translate the exact sequence of instructions into a computer language and now have a program which can find 7% of any number.
If however, I thought the correct method was to divide N by seven and multiply the result by 100 and program it accordingly, the computer will simply come up with the wrong answer faster than I could. Computers have absolutely no sense, common or otherwise. All they do is blindly obey instructions input to them by humans. You can run the same faulty program on a bank of Raptor class supercomputers and they will still come up with the wrong answer.
If you fundamentally do not know how to solve a problem, no computer, no matter how powerful, can help you. This is precisely the situation the Met Office are in.
Pointman
http://thepointman.wordpress.com/
The Met Office wanting new and better computers reminds of 10 year old boys who swear that with their new sneakers they can race across the parking lot faster. We would all smile and think they’re (Met Office) cute except for the devastating economy/human progress thing.
McLean et al. 2009 paper. Non-AGU have to pay up.
http://www.agu.org/journals/ABS/2009/2008JD011637.shtml
The power of that pattern matching computer in Piers head is truly amazing.
I don’t know how the Met office can stand it. Constantly being shown up by one guy with a laptop. Now if I were in their shoes I’d hire him and see what he and his algorithms can do with some serious computing power.
Not much bang for the buck. Is it wishful thinking, repetition or a script that guides the MET office forecast.
Compare it to a Stradivarius violin. Anyone with enough money can own one but you cannot wish it to produce good music and not everyone can play it properly. If however, the right violinist is hired to play it the music will be more than worth the cost.
Reply to orkneygal, the answer is no, General Eisenhower did not base his decision to go on wishful thinking. And no one would have dared give him that kind of forecast. And they had a lot less to work with except their knowledge, work and integrity.
Reply to WyattH, RPMs, its not necessarily the actor but the script writer. Although, the actor probably would not notice or care. His or her job is to deliver the lines in a convincing manner. RPMs less or more than an hour.
It would be nice if Brian mentioned that he was doing spreadsheets for Kristen Byrnes comparing SOI and global temps years before the McLean paper
orkneygal says:
January 12, 2011 at 3:21 am
Is this the same British weather office that General Eisenhauer relied upon to send the Soldiers of Democracy across the Channel into Normandy to begin the liberation of Europe?
He had three weather groups, the MET, the US Army Air Corps and the Navy, reporting to Group Captain J.M. Stagg who briefed him.
With respect to Piers et al and recognising their need to earn a crust, they do need to publish their methodology if they are to be Scientifically credible.
If not their methodology, then at least all their forcasts.
As it stands, we hear of their successes but possibly not their failures.
The Met Office often gets it right that is taken by many as an expectation and a given but they really cop the flack when they get it wrong.
I suggest that they (Piers et al) supply a daily, monthly and quartely forecast under an NDA to a disinterested stakeholder for release a rolling year later so that a true evaluation of the worth of their methodology can be established.
I seem to have irked a number of people.
To all you folk eager have it known that Piers Corbyn’s forecasts are significantly more skilfull than the Met Office’s: I have actually paid for his services myself in the past. If you have done the same then I will be pleased to hear about it. As for the self-published testimonials, all I can say is I’m a skeptic. I want extraordinary evidence to back extraordinary claims, no matter who is making them.
But my point was not to criticise maverick self-employed forecasters. My point is that establishing predictive skill takes evidence than Jo Nova hasn’t presented. Her claim is premature, outright hogwash, or there is a genuinely interesting story still not being told. I hope it’s the latter.
Vuk etc. says:
January 12, 2011 at 3:46 am
Lucy Skywalker says:
…………….
Could help if you monitor Stratosphere. First serious ‘break in’ of this winter happened 2 days ago. What does it mean? No idea.
http://www.cpc.ncep.noaa.gov/products/stratosphere/temperature/05mb9065.gif
That is called sudden stratospheric warming:
http://www.google.ca/search?q=sudden+stratospheric+warming+&hl=en&sa=2
Ther is evidence that Coronal Mass Ejection’s (CME) cause them. Piers Corbyn showed one last year on his site that reversed the polar vortex. There was one that hit earth earlier this week and another one forcast for january 14th. Bottom left of this page:
http://spaceweather.com/
I visit the Weather action website occasionally, but I really don’t know what to make of Piers. He is an eccentric with an apparently unique technique for forecasting weather.
I would love it though if he got sponsorship and was able to make everything public. His ‘technique’ would seem to be an obvious candidate for a research grant. I mean wouldn’t be great if a forecaster could tell the people of Brisbane they were going to be flooded years before the floods arrive!