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:
The likes of Mr Telford are increasingly, or should be increasingly, reviewing their positions. Why? well the problem for Gavin et al is that ever so gradually the big institutions appear to be “coming out”. This leaves the problem of judging the right moment for the individuals because if they leave it too late criminal procedings could be brought against them to save the rear ends of the directors of the larger instutions and the politicians. A politician in a tight scam is worse than a cornered rat.
Wet Orifice. Name says it all, really
I submitted the following comment on the Bishop Hill site this morning:
[Note: Doug Keenan’s article commented on the official UK government’s response to the repeated question both he and Lord Donoghue wanted the answer to, and included at one point the observation, “Most of the third paragraph is verbiage.” The reader should read that paragraph of the government response, to understand what the following comment, by me, is proceeding from.]
It is easy to cut through that verbiage: That third paragraph is saying, “the science is settled”.
I have shown, 2 1/2 years ago now, that the science is NOT settled, but wrong and indeed incompetent. As I have said over and over, climate science will not advance until my definitive evidence is properly confronted, and accepted by all. The finding presented here, that there is no statistically “significant temperature rise”, is tacit endorsement of my position, which first and foremost affirms that the Standard Atmosphere model is the stable, UNCHANGING (except to changes in the incident solar power), governing equilibrium state of the troposphere. It is that simple, and it is a revolutionary disproof of consensus climate science. In my Venus/Earth comparison, I found that the Standard Atmosphere for Earth differs from the actual temperature vs pressure profile of Venus (over the range of Earth tropospheric pressures) only due to the difference in solar distance of the two planets (simply put, and for the 1,000 mb pressure level in particular: The Venus temperature at 1000 mb is exactly what the surface temperature of Earth–also at 1000 mb–would be, if Earth were as close to the Sun as Venus–even though Venus has over 2400 times the CO2 level as Earth, 96.5% vs. .04%.) There is NO CO2 “greenhouse effect”, of increasing temperature with increasing CO2–and thus there are NO competent climate scientists, who all promulgate that false and incompetent theory.
@Robin:
Per:
Well, that explains a great deal… “The Limits to Growth” was complete and utter trash that started the “project log problems with computers” bunk in the first place. Now we see why they are so negative.
For me, and most other of the Skeptics I’ve met, we’re perfectly fine at present. Not feeling at all out of balance, nor needing any new “spiritual, ethical, …” whatever “emptiness” in our lives. So don’t need anyone “restoring within us” their POV…
But at least now we know why the Warmers are such Grumpy Gusses and pushing for all that “social, political” etc. force fed change. They are morally bankrupt and philosophically empty (per that quote) and desperately seeking a “fill up”… Perhaps they ought to be delivered to hospital for an enema… at least then they would be feeling “restored within”…
Nick Stokes says:
May 27, 2013 at 11:48 am
The matter is very simple. Keenan’s statistic assumes no trend. As always with Alarmists, the Met Office’s statistic assumes a trend. Keenan’s gives a thousand times better fit to the data.
The Bishop has created his own short version for laymen. Take a look.
Second that! The BH post is well worth the read.
Gary Pearse says: May 27, 2013 at 12:44 pm
Gary, you quoted as the supposed Met Office admission, this from Keenan:
““Doug McNeall is a statistician. He and I have had cordial e-mail discussions in the past. In particular, after my op-ed piece in WSJ appeared, on 12 August 2011, McNeall sent me an e-mail stating that the trending autoregressive model (used by the Met Office) is “simply inadequate””
If you go to the email which DK linked, what McNeall actually said was:
“A linear trend, while conveniently easy to understand and apply, is simply inadequate to capture all of the timescales that are apparent in the Earth system.”
And that isn’t news to anyone.
Snotrocket says: May 27, 2013 at 1:05 pm
“you quoted Lucia – and then, didn’t say whether you agree with her or not.”
I referred you to her statement – of course I agree with it. And yes, the confused question from a Lord was written, not verbal.
Does this mean our children will know what snow is after all? : )
lsvalgaard says:
(Cloud cover for Northern Europe has been reconstructed back to AD 1000.)
http://www.leif.org/research/Norway-Cloud-cover-Reconstr.png
“In particular, the Maunder Minimum had low cloud cover [should be high temps]”
With winter temp’s up there being far more variable than in summer, surely less cloud in winter means colder?
The UK Met Office would be better off with some seaweed and a old pine cone to make their long range “Predictions”
Nick Stokes says:
May 27, 2013 at 1:27 pm
…….
I don’t know where you live, but most of ‘normal and sane’ people who live in UK, don’t really care what Met Office admits or doesn’t; people have realised for some time now that their projections of so called ‘global warming’ now transmuted into climate change, and consequently long term forecasts have become a national joke.
Hey Nick! You must have heard: ‘When you’re in a hole, stop digging’?
So, what is it? A ‘confused discussion’ (your earlier comment) or a ‘confused question’ (your latest post).
And why did you need to be pushed into agreeing with Lucia’s quote? ‘Cos I could argue you took her out of context. She was discussing Mark Bofill’s earlier comment – my bold (‘I understand Lucia has argued that it isn’t possible that weather noise obeys this statistical model?’) So perhaps it was not so much an argument about ARIMA being wrong for use in climate stats as being wrong in weather stats. (I guess you would argue that weather is climate/climate is weather – delete what does not apply)
Good post. Unfortunately those to whom you addressed will not see that.
None so blind as he who will not see.
Bishop Hill writes “the Met Office should now publicly withdraw the claim. That is, the Met Office should admit that the warming shown by the global-temperature record since 1880 (or indeed 1850) might be reasonably attributed to natural random variation.”
Maybe they should, but until they do it can hardly be claimed that “Met Office admits claims of significant temperature rise untenable”
Seems to me that we are still at the stage of Met Office SHOULD admit claim of significant temperature rise untenable.
Until they do this analyses will be looked at by many as just another denier rant.
0.8 degree….in 133 years
I’m no longer amazed that there are people out there that stupid
Theo Goodwin says: May 27, 2013 at 1:26 pm
“The matter is very simple. Keenan’s statistic assumes no trend. As always with Alarmists, the Met Office’s statistic assumes a trend. Keenan’s gives a thousand times better fit to the data.”
I think the matter is very simple. This post makes a very bold claim. “Met Office admits claims of significant temperature rise untenable”. And there is no basis for it, none even offerred. The Met office made no such admission.
So now it comes down to, well, maybe Keenan has a better trendless model. But no, the comparison cited is between a first order autoregressive model with trend, and a third order (“driftless”) model. That’s no fair comparison – the third order model has more parameters to play with. And as Lucia said, no-one expects that linear rise since 1850 is the expected result of AGW.
I’m with Richard Telford here, and I must say I have found it extremely disappointing that no-one here nor at Bishop Hill have explained what the ARIMA(3,1,0) thing is about. Just because there is a statistical model which fits the data better doesn’t mean that it is right, and even if it is right then what are its physical implications and future implications? It seems that I shall have to go to Lucia’s blackboard to find the answers, because no-one is able to elucidate here.
I agree that this result undermines some of the science behind AGW, but not all of it I think. After all, neither of the 2 models under consideration include solar forcings. Models with which to test statistical significance are just models, a decent but not perfect representation of the real world.
(Richard Telford wrote: The problem with Keenan’s analysis should be obvious. The 1 in the ARIMA(3,1,0) is needed because the data are non-stationary – the mean is not constant – the temperature is increasing. Whether this temperature increase is removed by differencing as Keenan has done, or fitted with a linear trend, depends on the aims of the analysis. Both agree that there is significant warming, and neither model can determine the cause of the warming.
As the climate forcings have not increased in a linear fashion, it is not surprising that a linear trend is not a very good fit to the instrumental data.)
Rich.
climatereason says:
May 27, 2013 at 8:43 am
I looked at that reconstruction of the CET to 1538 last week on an earlier thread here. What a tremendous store of information it holds within its records. I thought it interesting that the very beginning of your reconstruction, at 1538, shows a tailing off from a warm period with a strong warming. That made me wonder what the previous years looked like, although I suspect that the warm trend would go back to early 1500s, perhaps to 1510. That would make the length of the warming approximately equal to the current warm trend from the late70s till around the late 2000s. Which could mean that you don’t have to go all the way back to the MWP to find a period that matches recent years for warming.
climatereason says:
May 27, 2013 at 8:43 am
I would also like to say thanks for sharing your work where others can appreciate it.
At the 95% ‘confidence’ level, green jelly beans cause acne. (Not the other 19 colors checked.) xkcd proved it.
Make that Manhattan ==> New Jersey.
ARIMA(3,1,0) means that the time series is non-stationary which means there is a trend. An upward trend, caused probably be greenhouse gases.
Nick Stokes says: May 27, 2013 at 12:40 pm
Louis says: May 27, 2013 at 12:28 pm
Louis, it does not answer the question. This is not the Met Office “admits claims of significant temperature rise untenable”. It is still only the case that Douglas Keenan thinks they are untenable. The Met Office didn’t say so.
The Met Office finally admitted (after 6 attempts to get them to answer) that the driftless model is 1000 times more likely to fit the data than the statistical model they use. The driftless model says that the .8 C of warming since 1850 is well within what could be expected from natural variations in temperature, thus it is not significant. So to any honest person, this is an indirect admission from the Met Office that their claim of significant temperature rise cannot be defended using current data and is thus “untenable.”
Now, if you want to argue that given more time the data will change, that’s your prerogative. But then it becomes a matter of faith rather than science. The facts are, current data do not show significant warming, the small warming trend we recently experienced has stopped for over a decade while CO2 continues to increase, and there is no evidence for climate doom or other predicted ill-effects of warming beyond natural variability. For now, Occam’s razor requires us to lead with the simplest explanation that explains the current facts. When and if the data change, then we can revisit the conclusion. Until then, the simplest explanation stands.
Ulric Lyons says:
May 27, 2013 at 1:31 pm
“In particular, the Maunder Minimum had low cloud cover [should be high temps]”
With winter temp’s up there being far more variable than in summer, surely less cloud in winter means colder?
So, you admit that cloud cover was low during the MM.
Snotrocket says:
May 27, 2013 at 1:47 pm
Hey Nick! You must have heard: ‘When you’re in a hole, stop digging’?
So, what is it? A ‘confused discussion’ (your earlier comment) or a ‘confused question’ (your latest post).
And why did you need to be pushed into agreeing with Lucia’s quote? ‘Cos I could argue you took her out of context. She was discussing Mark Bofill’s earlier comment – my bold (‘I understand Lucia has argued that it isn’t possible that weather noise obeys this statistical model?’) So perhaps it was not so much an argument about ARIMA being wrong for use in climate stats as being wrong in weather stats. (I guess you would argue that weather is climate/climate is weather – delete what does not apply)
———
Yikes! Don’t quote me as if I know what I’m talking about. I was mostly talking about the fact that I don’t understand ARIMA at all. Well, not much anyway. But particularly don’t attach significance to the weather / climate difference there; I glanced at an old blog post at the Blackboard and without a whole lot of thought or concern I came out with ‘weather noise’.
So far regarding the application here, I gather that:
1. It’s a better model as far as matching the data is concerned.
2. There’s an argument against it’s use pertaining to what it means physically, in reality, having to do with random walks. That and the fact that forcings shouldn’t cause a linear trend in temps. I don’t understand why yet, still trying to figure ARIMA out.