Excerpt from Bishop Hill (plus a cartoon from Josh) showing that the claim of a statistically significant temperature rise can’t be supported, and the Met office is ducking parliamentary questions: (h/t Randy Hughes)
Met Office admits claims of significant temperature rise untenable
This is a guest post by Doug Keenan.
It has been widely claimed that the increase in global temperatures since the late 1800s is too large to be reasonably attributed to natural random variation. Moreover, that claim is arguably the biggest reason for concern about global warming. The basis for the claim has recently been discussed in the UK Parliament. It turns out that the claim has no basis, and scientists at the Met Office have been trying to cover that up.
The Parliamentary Question that started this was put by Lord Donoughue on 8 November 2012. The Question is as follows.
To ask Her Majesty’s Government … whether they consider a rise in global temperature of 0.8 degrees Celsius since 1880 to be significant. [HL3050]
The Answer claimed that “the temperature rise since about 1880 is statistically significant”. This means that the temperature rise could not be reasonably attributed to natural random variation — i.e. global warming is real.
…
The issue here is the claim that “the temperature rise since about 1880 is statistically significant”, which was made by the Met Office in response to the original Question (HL3050). The basis for that claim has now been effectively acknowledged to be untenable. Possibly there is some other basis for the claim, but that seems extremely implausible: the claim does not seem to have any valid basis.
Go read the entire essay here: http://www.bishop-hill.net/blog/2013/5/27/met-office-admits-claims-of-significant-temperature-rise-unt.html
Josh has a go at them:
Jimbo says:
“Meanwhile in Germany it looks like they are having the coldest Spring in 40 years.”
I can relate. In my 45 years in North Carolina I cannot remember a spring this cold. This morning felt more like Easter than late May. Even the real old-timers I’ve talked to can’t remember it being this cold this late, but we did have a warm winter (so there ya go, global warming … er … uh … climate change /sarc).
@Doug Proctor – It should be noted that it’s much easier to make money out of a “single known cause than “many diffuse ones”.
Did I say that,,, really???
YES!
Is it true that the Met Office tried to buy Apple #1 as their next super computer?
Andrew, now might be the time to at least begin thinking about the future of WUWT after the impending collapse of CAGW. The first task that comes to mind is making arrangements for a complete and permanent archive of all activity on this site. The second that comes to mind is that it might be time to start examining other areas where pseudoscience in service of the social engineers is imposing significant social costs.
Weird that the Met Office would stick so tenaciously to a prediction when the accuracy of the easiest and most practiced thing they predict (weather, you know) is considered to be 80% today, 60% tomorrow, 40% the day after, and only 20% thereafter.
And since climate = weather integrated over the next 30 years (more or less), wouldn’t they have to stick with a 20% confidence level when predicting climate, since the first 3 days of that 30-year period are completely insignificant compared to the rest?
I won’t hold my breath for an answer.
Doug Proctor says:
“We have a non-unique solution type problem. The warmists have reframed it as a unique solution problem, and since they have found “a” solution, i.e. A-CO2, they have found THE solution. The post’s statistical statement is that the rise global temps since 1850 is not a unique solution situation. That is important and signficant in terms of the debate, but that is all: CO2 is sufficient but not necessary.”
I made the mistake of saying that @ur momisugly RealClimate once. You can probably imagine the reaction. Of course you are right, there’s way more unknowns than equations with which to isolate and solve for the variables. I will admit I can kind of understand working on solving (let’s say) 25 equations with 100 unknowns and finally finding “a” solution how one might think of it as “the” solution. That’s why a real consensus (to me) is one in which those closest to the hypothesis or “proposed solution” have convinced the rest of the scientific community of its validity; this is what climate science has utterly failed at doing.
There won’t be much support for the Met Office from the other side of the Atlantic considering upstate New York just got 3 FEET of that “stuff we’re never supposed to see ever again”:
http://www.foxnews.com/weather/2013/05/27/where-spring-upstate-new-york-gets-3-feet-snow-on-memorial-day-weekend/?test=latestnews
Or was “summer” just a few weeks long?
Lord Donoughue –
kim says:
May 27, 2013 at 7:43 am
Who is he [and more]:
A link [the peerless, perennially permanently pluperfect, Wikipedia, but a start] –
http://en.wikipedia.org/wiki/Bernard_Donoughue,_Baron_Donoughue
Paraphrasing, a lad who made good, ‘Economist’ journalist [when the E was a decent rag, I guess]; a senior adviser in the 1970s to Prime Minister Harold Wilson [the man with more faces than the Town Hall clock] and Jim Callaghan [1974-79], then back to the Economist, then ‘The Times’, whence he was apparently dismissed by Rupert Murdoch. [Is that a badge of honour in some circles?]. Later, various good works – Orchestras, Betting Commissions (the Starting Price Regulatory Commission (SPRC)), etc. then a junior minister under ‘Tony Blair’ as we call Tony B. Liar.
A couple of other comments if I may: –
In London – and most of the UK, I think, but certainly London – the ‘Met’ is the Metropolitan Police. The Met Office is the Met Office.
Now that Parliament has been told there is no – ahhhhh, ummm, whatever the latest ‘in-phrase’ is – Non-Natural Hotness-Spread, say, can we get back to trying to cut the deficit [if eliminated by raising income tax n the UK it would need an increase of 26% in all income tax rates to eliminate], and defending our islands.
Yes, I’m British, in London.
And can we have our money back?
Please?
AlecM says: May 27, 2013 at 9:14 am
The key issue is that Climate Alchemy depends upon incorrect physics from Sagan and Houghton. Hansen codified it and in 2011 was forced to claim that aerosol cooling was exactly equal and opposite GHG-AGW, and still the lunatics believed in the religion.
////////////////////////////////////////
Alec
They had to argue that aerosol cooling was more than opposite GHG-AGW.
Manmade CO2 emmissions only began to rapidly increase as from WW2. There was a very rapid increase in CO2 emissions between say 1940 and 1970 when compared to the rate of increase between say 1900 and 1930.
However just as manmade CO2 emissions rapidly increased temperatures fell. To explain the fall in temperatures between 1940 to say mid 1970s required the ‘warmists’ to argue that increases in manmade aerosol emissions had a greater than equal and opposite effect to the rise in GHG-AGW and that is why instead of observing warming as from the 1940s, there was cooling post 1940s.
It was only after steps were taken to combat manmade aerosol emissions (claimed to be in the late 1970s) that one began to see the effect of global warming brought about by manmade emissions of CO2.
I’d like clarification as to what is meant by the word significant. Because it seems to me that it is being used in two ways.
On the one hand you could defend the claim that temperature has risen risen significantly when tested against the null hypothesis that it stayed the same or decreased.
On the other hand it is certainly false that the temperature has risen significantly when tested against the null hypothesis that the climate is varying naturally.
William Astley says:
May 27, 2013 at 8:55 am
Hint solar modulation of clouds.
there is no evidence for that, quite the contrary, e.g. http://www.leif.org/EOS/Cloud-Cover-GCR-Disconnect.png
lsvalgaard says:
May 27, 2013 at 10:47 am
———————————–
Correlation not too bad until about 2004. Statistically significant? Maybe.
What does it look like on a centennial or millennial scale instead of decennial? Thanks.
@richard telford
‘The salary of academics is not dependent on their opinions – a concept known as academic freedom’
H’mm
But simple salary is not the sole component of the reward package that academia offers to its participants. If it were then few would choose to do the job…there are bigger pickings for tolerably bright kids elsewhere.
Academia also values direct and indirect recognition among one’s peers and others. Just look at the CV of a moderately successful Professor and it will be littered with medals and prizes and other baubles and honours that a city trader would consider totally irrelevant – but matter a lot in that field. There are conference keynotes, books to write, appearances to make, and – for example – the invitation to become an IPCC Author. And of course, all academics aspire to promotion within their field.
None of these are salary dependent, but you be assured that they come few and far between within climatology to those who hold differing views from ‘the consensus’
It is disingenuous to pretend that because the direct salary is not opinion dependent an academic is under no pressure to conform. It is a group like any other group ..and conformity is prized.
Keenan’s analysis highlights something Richard Courtney and I have been saying for some time on these boards: to claim “statistical significance”, you must use a model of the stochastic process. And, if your model is no good, your estimation of statistical significance is bogus.
Far too often, particularly in the climate sciences, I see statistics abused in the service of denying what can be seen with the naked eye. In private industry, where we have to be sure of our results or there may be catastrophic consequences, we would never use a process model which disagreed with what we could see with a cursory glance.
It has been very frustrating to me, the challenges I have received along these lines. People assume that a painfully overwrought analysis carries more weight than simple inspection. That, if you can dress your argument up in terms of Hurst parameters, or ARIMA processes, or Fokker-Plank equations, or what have you, then it is perforce of greater validity than simply looking at the data and remarking upon what you see. Nothing could be further from the truth. It is painfully easy to devise sophisticated mathematical arguments that the emperor’s clothes are substantial, when a simple observation indicates that he is starkers.
In-depth analysis with powerful mathematics has its place. But, in industry, we strive for what we call “sanity checks”, simple relationships which must be verified in order to proclaim the sophisticated models valid. Very much along the lines of Doug Proctor @ur momisugly May 27, 2013 at 9:59 am: you must satisfy necessary conditions, not merely sufficient.
@Ian H
A true null hypothesis is that there has been no change. The null does not make an assumption about cause and effect. Significance testing sets a definable measure of the chance that a phenomenon differs from that null hypothesis. I would expect everyone on this site to agree that there has been significant warming since 1850, the CET shows it clearly.
The 0.05 significance level sets a comparable bar on chance. Many areas of science go well beyond this level, to 0.01 and even 0.001. I hope this clarifies matters.
AlecM
So, you do not get any energy transfer until the radiation fields interact destructively. The result of this is that ‘back radiation’ is an aborted foetus of science.
Along this line I have wondered how the famous 2-slit quantum experiment applies to this radiation. The experiment demonstrates that light travelling through 2 slits forms an interference pattern on a screen. In other words, radiation is carried via waves. So, when we have two sources of radiation heading towards each other what exactly happens? Does the energy just disappear?
On the other side we have conservation of energy requirements. Or, is all this just phantom energy and not real.
Maybe you could clear up how this all works out.
Richd. Telford: “If you have to stoop to using ad hominem arguments, …”
Difficult to call Telford’s original comment “argument;” but, surely it is an ad hominem attack on Douglas Keenan. There’s no science, or maths, in it. Seemed like ‘stooping’ to me.
John West says:
May 27, 2013 at 10:18 am
Thanks for trying. Still, it is good to point out from time to time that “we can’t think of any other way it could happen” is not a statement of proof, but a remarkably frank admission of personal limitations.
Douglas J. Keenan says:
May 27, 2013 at 8:31 am
Thank you, Mr. Keenan, for your brilliant and heroic efforts in behalf of science.
Terence Mills published an editorial review article in Climatic Change in 2010 using structural time series models to show that a driftless I(1) process beat the stationary trend model.
Mills T. C. (2010) Skinning a cat: alternative models of representing temperature trends. Climatic Change 101: 415-426, DOI 10.1007/s10584-010-9801-1.
http://link.springer.com/article/10.1007%2Fs10584-010-9801-1?LI=true
While he acknowledged that you could plausibly adopt a number of different statistical models that imply different things about the underlying GW trend, the model he finds has the most support treats the trend portion as a driftless random walk. He concludes:
Although Terence won’t be at my conference next week, a number of the main players in the debate over persistence and nonstationarity in temperature data will be presenting their work and debating these very topics.
http://econapps-in-climatology.webs.com/
The IPCC should have devoted a whole chapter to the stationarity question long ago, but unfortunately it’s way over the heads of the LA’s and they have remained mired in their naive linear trend+AR(1) model for years.
I’m always a bit doubtful about scientific conclusions drawn in a politcal arena…
John Tillman says:
May 27, 2013 at 11:03 am
Correlation not too bad until about 2004. Statistically significant? Maybe.
This is a typical example of a correlation that looked good when it was publicized by Svensmark and company. A mark of a real connection would be that the correlation continued when new data is added. A mark of a spurious connection is that the correlations fails when new data is added.
What does it look like on a centennial or millennial scale instead of decennial?
There is no data on cloud cover on longer time scales.
A.D. Everard says:
May 27, 2013 at 9:35 am
… Let’s see what Parliament does with it. I guess they’ve got some thinking to do.
They will, no doubt, look vacantly at lobbyists who will tell them what to do.
Cloud cover for Northern Europe has been reconstructed back to AD 1000. Don’t know about other regions, nor how reliable data divined from the dreaded dendro might be.
ftp://ftp.ncdc.noaa.gov/pub/data/paleo/treering/isotope/europe/norway/forfjorddalen2012.txt
The correlation does certainly appear to have failed for the past decade.
Thanks.
Dr. McKitrick:
You might think that a test of statistical significance would be required before pursuing a policy of dismantling industrial civilization based upon the untested findings. But apparently too many salaries & grants depend on practicing voodoo “science” corrupted by government & foundation funding.
IMO, the IPCC won’t run such a test until the governments funding it require this elementary exercise. Governments won’t do so until taxpayers & their representatives demand it. Maybe you can convince the Harper government to make Canada the leader in this movement toward scientific validity.
Thanks for all you’ve done.