[Update, March 3: Lindzen’s original 2008 chart showed the HadCRU surface record, not the UAH satellite record. The piece has been corrected accordingly. ]
As I predicted two weeks ago, there has been a media and blog frenzy over the statement by climatologist Phil Jones that global warming since 1995 has not been “statistically significant”.
Now might be a good time to reflect on the origins of this latest variation on the “global warming has stopped” meme. And it turns out that the originator of the “no significant warming since 1995” variation is none other than climate contrarian Richard Lindzen, ably assisted by blogger Anthony Watts (of WattsUpWithThat fame) and physicist Lubos Motl. Indeed, the history of this meme points up a disturbing symbiosis between the U.K. media and the contrarian blogs.
Can I call them or what? Back on February 16, I wrote:
Get ready. Lies originating in the U.K. over the weekend in newspaper stories by Jonathan Leake of the Times and Jonathan Petre of the Mail on Sunday, are about to hit the contrarian echo chamber.
As I write this, Google counts 2.3 million hits for the search with the exact terms “warming since 1995” and “Phil Jones”.
Early media outlets onto the bandwagon included the usual suspects, FoxNews and the Wall Street Journal, with the Washington Post’s George Will weighing a few days later. Here in Canada, contrarians were almost as quick – the Jones 1995 statement was used by the National Post’s Lorne Gunter (explaining the meaning of ClimateGate) and the Globe and Mail’s Margaret Wente (demonstrating that the “science isn’t settled”). And, as we saw recently, Gunter has been double dipping, recycling the bogus claim to attack the credibility of the instrumental temperature record from HadCRU and NASA.
Indeed, there has been so much media commentary about Jones’s supposed “U-turn” that Nigel Lawson‘s Global Warming Policy Foundation has reprinted no fewer than four opinion pieces touching on the subject in the last week alone.
The sequence of events from the BBC “online interview” with Jones seemed straightforward enough, as I outlined in my previous post, Morano sends lies from UK Times and Daily Mail around the world.
- Phil Jones tells the BBC that the global temperature trend from 1995-2009 is “positive, but not significant at the 95% significance level”, although “quite close”.
- Daily Mail publishes misleading article, with headline “Phil Jones: no warming since 1995”.
- WattsUpWithThat and Marc Morano point to Daily Mail article.
- Media and blog explosion ensues.
But a nagging question remained. The BBC’s Roger Harrabin stated that his questions were “gathered from sceptics” and indeed many of them were familiar to me as thinly disguised contrarian talking points. For example, some time ago I wrote about the Ian Plimer’s invocation of previous warming periods to argue that modern warming might also have the same natural causes.
Yet I had never come across the particular claim of lack of “statistically significant warming” since 1995. Presumably, though, this particular contrarian talking point was already out in the denialosphere. And, indeed, a quick check of Google turned up the earliest known sighting of this version of the “global warming has slowed down” meme – a short and sweet email from Richard Lindzen to Anthony Watts in March 2008.
Look at the attached. There has been no warming since 1997 and no statistically significant warming since 1995. Why bother with the arguments about an El Nino anomaly in 1998? (Incidentally, the red fuzz represents the error ‘bars’.)
The supplied chart was from the HadCRU record:
The theme was revisited again last December by physicist (and climate contrarian) Lubos Motl, who used UAH satellite data to conclude:
We can only say that it is “somewhat more likely than not” that the underlying trend in 1995-2009 was a warming trend rather than a cooling trend. Saying that the warming since 1995 was “very likely” is already way too ambitious a goal that the data don’t support.
We’ll return to this analysis in an upcoming post. For now, I will note that, once again, Anthony Watts was there to broadcast Motl’s analysis in a Boxing Day post. Now it’s not clear if Harrabin got this question directly from Motl (via Watts), ensuing blog commentary (it even appeared in one BBC blog comment), or from some other source, say, a willing think tank. (In this connection, it’s worth noting that Ian Plimer and Richard Lindzen are both advisors for the aforementioned Global Warming Policy Federation).
Whatever the exact path of the the contrarian talking point into Harrabin’s consciousness, it seems fair to update the history of this meme:
- October, 2008: Lindzen-Watts
- December, 2009: Motl-Watts (again!)
- February 13, 2010: Jones-BBC
- February 14, 2010: Daily Mail: “no warming since 1995”
- February 15, 2010: Watts (for the third time!)-Morano
The end result is that an irrelevant talking point about lack of warming, hitherto propounded by discredited contrarians, has now become an “admission” from Phil Jones about of lack of warming in HadCRU. That’s what happens when confused journalists like Harrabin get their science from blogs, and other even more gullible and biased journalists like Jonathan Petre run with it.
So why did Lindzen and Motl choose 1995 as their starting point? In a comment on Deltoid’s post on the Economist calling the Daily Mail to account (calling the lie for what it is is), I engaged Motl on this very point:
How about since 1994? Cherry-picking much?
Motl answered in the following comment, rather helpfully confirming the cherry-picking:
Dear Deep Climate,
your point is somewhat shallow. In this thread, I discussed 1995 as opposed to 1994 because that’s the year that BBC asked Phil Jones about, and for a good reason. 1995 is the earliest year when the statistical significance of the trend from that year to 2009 safely fails. Since 1994, you could get a technically significant trend. It would still not be a robust result because a small change of the beginning year would destroy the statistical significance …
BBC asked a very good and highly relevant question in a technically accurate fashion and Jones had to answer it – the answer was clearly “Yes”. He was also allowed to add refinements. It’s just the religious bigots like most of the readers above who dislike the question and the answer – because the answer, “Yes, there’s been no statistically significant trend for 15 years”, shows that even if there’s anything such as “global warming”, it’s not really worth talking about. And you don’t like this fact for the same reason why an Islamic bigot doesn’t like the picture of Mohammed as just another obsessed warrior and/or terrorist: because it’s an inconvenient truth.
That answer rings a little hollow, doesn’t it – now that we know the cheerful Lubos likely played a key role in getting the question asked in the first place!
And that’s not the end of the story – for it seems that even the misleading, cherrypicked claims of “no statistically significant warming since 1995” are not quite right – at least not for the surface global temperature record. But that’s a tale for another time – stay tuned.
Here is another … interesting example of Motl-logic in that Deltoid thread.
That’s the case of the warming since 1995, too. From the viewpoint of statistics, it doesn’t exist. Building on a statistically insignificant trend would be exactly as illegitimate as building on a non-existent trend. So the statement that Jones said that there was no global warming since 1995 is not literally accurate, but it is surely more correct than the statement that he didn’t say that there was no global warming since 1995.
Sounds like a conspiracy.
Not really – all these statements, as misleading as they were, were out in the open. There was gullibility on the part of the BBC, however.
The question was certatinly framed by a contrarian. They chose the earliest year in a particular data set that would result in the trend falling just short of the 95% confidence level, so they can generate the right talking points.
I see Motl admits the same.
Motl further spins the results…
“I am not sure whether any of you understands basic science terminology, or whether you want to understand it, but if an observation extracted from the data is “not statistically significant”, then it means that it’s plausible or likely that it’s fundamentally equal to zero and the observation was caused just by noise.”
So Motl believes (or pretends to) that 5-10% is the equivalent as plausible (>50%) or likely (>66%). There’s a reason why these folks aren’t taken seriously among scientists. They are good at working the media, however.
Wow, what a lovely post by Motl (sarc). Really? I can’t believe the invective and rhetoric.
That analysis applies to the CRU, and probably the UAH data. I am not sure what the stat. sig. is for the warming in GISS or RSS.
Referring to statistical significance is a clever tractic being used by thos ein denial to further confuse the public. First it was, it has been cooling, then that change to there has been no warming, and now they are scarping the barrle with there has been no warming at the 95% confidence level in the coolest of all the four global SAT data sets. The year 1995 is clearly cherry picked. Tamino has posted excellent articles on the matter of how long a record one needs for stat. sig.; for the GISS one has to go back to 1996.
Tamino also observes that:
“Let’s say that less than 15 years of data allows no confident conclusion about whether the trend in GISS data is warming or cooling. That does not mean that there’s been no warming trend in those 15 years — or in the last 10, or 9, or 8, or 7, or 6 years, or three and a half days. It only means that the trend cannot be established with statistical significance.”
DC, those in denial will be playing this game of cherry picking and shifting the goal posts in 2040.
I love how the denialists in one breath claim (falsely) that CRU fudged the warming trend in HadCRUT temperature record data, and in the next breath claim that the 15-yr warming in said data set is not statistically significant at the 95% confidence level. It is truly absurd.
Talking of which Accuweather’s Bastardi is really indulging in some creative (and truly desperate) spinning in trying to explain why global tropospheric temperatures are at record (or near record) highs. He even floated the idea that the satellite might be malfunctioning…well maybe it is, but the GISS, RSS and RATPAC data corroborate its findings, so that is wishful thinking on his part.
What on earth will they do when their beloved (and suspect) UAH MSU data start failing them?
Interesting how this one goes in circles. Starts in the blogs, goes into the news and then the blogs report the “news”.
This is nearly identical to a tactic that the Bush administration used repeatedly:
Administration official “A” mentions something on background to the media. Media repeat it without attribution. Administration official “B” then says on Meet the Press, “Well, according to the New York Times….”
[DC: It is similar – but in this case, the PR deception is even more insidious. The contrarians have managed to get an actual mainstream climate scientist to supposedly “admit” one of their cherished talking points. ]
It was obvious from the start that the question was designed to ambush Phil Jones and unfortunately he wasn’t on the ball enough to give a better answer.
He could have explained that, in order to distinguish a genuine trend in global temperatures from natural interannual variability, we have to look at more than 15 years of data. As the normal solar cycle lasts around 11 years, any single 15-year span of temperatures could start at a maximum of solar activity and end at a minimum, or vice versa, thus giving us a misleadingly high or low trend in global temperatures. This could easily make the difference between finding a warming trend of 0.1C per decade or 0.3C per decade – a factor of 3 between them. Choosing 22-year trends smooths out some of the effects of the solar cycle and give us a better chance of distinguishing a long-term trend from short-term natural variability. When we do this, we see clear, consistent and statistically significant warming of around 0.18C per decade.
I created a web page to try to show how choosing appropriate time periods is necessary to try to distinguish a trend from the ‘noise’ of natural variability –
Climate scientists have to try not to be caught out by questions and comments based on such obvious cherry-picking of data.
“What on earth will they do when their beloved (and suspect) UAH MSU data start failing them?”
It will only confirm to them the true breadth of the *conspiracy*; even Roy Spencer can be corrupted! Is nobody immune from the evil spell of the *warmists*??!!
It’s only a matter of time before Spencer gets thrown under the bus.
Snap! I just discovered the same thing, with an additional publication in Australian rightwing rag, Quadrant.
We are constantly hectored as to how important the peer-review process is and how we must defer to it. Yet, Phil Jones, when asked at the UK inquiry how often scientists had asked to see his raw data, see details of his methodology or review his computer code, answered:
“They’ve never asked.”
Which simply means you don’t understand what reviewers are looking for when they review.
I love the way that denialists have become the instant arbitrators of how they think science works, and when they find out they don’t know how science works, insists it *should* work the way they previously imagined it worked.
Science has been doing fine without your help for a very long time, Paul S.
Paul S, peer review does not entail re-analyzing the raw data. The important question is whether the data put forth by the author support the conclusions made by the author. The methods are included in the paper and are supposed to be sufficient for someone skilled in the art (sorry – patentspeak) to replicate the results.
Reviewers are supposed to know the field well enough to evaluate whether a given manuscript, including the background, data, data analysis, and conclusions, is worth publishing and whether the paper adds something to the science.
I am not sure how to interpret your post. Scientists do not typically ask other scientists to share code. That is not how we independently verify the validity of others’ research. We do it ourselves, and if both party’s got it right then we get the same answer. If not, then at that point we might share. There is a move now to publish the code at the same time as the paper as was done by Mann et al. (2009).
As for peer review. I suggest you submit a paper for peer review, it is an exhaustive and sometimes humbling experience.
And papers are required to discuss the data and provide details on the methodology. If a paper is well written others should be able to reproduce the results without the author’s help. I don’t get your point.
Thanks for posting that link – it’s an excellent visual demonstration of the need to consider longer term trends.
I can just imagine what went through the mind of whoever had the “statistical significance” brainwave –
“Hey, let’s calculate the minimum number of years to give a statistically significant warming trend (let’s call it y) and then point out there has been no statistically significant warming for y-1 years. The beauty is we can do this every year – we have abolished global warming for ever.
Dear Deep Climate, I wrote a kind of response to this text:
All the best
[DC: It turns out I’m “promoting panic about AGW”. Oh well, at least I’m no longer of the same mindset as “an Islamic bigot”.
Apart from that, it’s a fascinating case study in how to move goal posts.
Equally fascinating is Lumo’s take on the dangers posed by AGW “alarmists”.
The comments/observations that I made about Lindzen also apply to you. Oh, and regarding your rant posted here by DC. Now that is a great example of alarmism and paranoia on your part. I just hope that the truth does not continue to elude you for too much longer. In the meantime, please try and not embark on knowingly using dodgy statistical analysis to support your denialism.
That’s one of the most reprehensible efforts to exploit the death of an obviously crazy couple and their innocent child I’ve ever seen.
In other words, typical for Motl.
[DC: I hesitated passing that on, but I think people need to realize just how despicable Motl is. Dealing with him been an eye-opening experience for me. I thank you and everyone else for showing extreme forbearance. ]
That excerpt on the Argentina case is one of the sickest bits of exploitative propaganda I have ever seen anywhere, period. DC, you are quite right to highlight it.
The chart that Lindzen had used to prove that there has not been a statistically significant warming is not from UAH, but from Jones’s own CRU data set! Certainly, UAH and RSS also show similar trends as CRU since 1995
[DC: Thanks, corrected above. ]
“It’s only a matter of time before Spencer gets thrown under the bus.”
The rumblings have already begun. Since this post: http://wattsupwiththat.com/2010/02/04/january-uah-global-temperature-warmest/ — and every thread since contains a comment or two about unreliable satellites and so forth.
They won’t even break stride. And really, they have a point: if all of the temperature records match up (and they do) then all of them are the product of a single global-spanning, centuries-long conspiracy. Or else they, you know, reflect a common reality in the physical world being measured by different means.
I blow the lid off the conspiracy here: http://theidiottracker.blogspot.com/2010/02/global-warming-hoax.html
[DC: Not sure why, but I had to rescue this from the spam filter – sorry for the delay. ]
What I neglected to mention the other day, and this is perhaps what really needs to be emphasized, is that Lindzen is not operating in good faith. He has divorced himself with the science and that email to A. Watts is scary, b/c that is not how a reputable scientists in the search for truth thinks.
This is LindzenGate. He is trying to hide the incline while at the same time his ilk are accusing the CRU of fudging the data. Here Lindzen is explaining to Watts how to fudge the data to avoid getting statistically significant warming.
As andrew pointed out above, they can apply this tactic ad infinitum and continue to deceive the masses.
I was dubious of Lindzen before, now I know for sure he is not sincere or impartial, and his work is probably riddled with confirmation bias (e.g., . Lindzen and Choi, 2009).
[DC: I would say “fudging” goes too far, but cherrypicking is a valid characterization. And he does seem to be more interested in propaganda than science. ]
Yes, I get a tad carried away. I agree that ‘cherrypicking’ may be more appropriate. Can you imagine the furor that would have erupted by the deniers had Jim Hansen sent such an email?
Perhaps I am wrong, but this simple email message from Lindzen is huge. He is telling someone how to manipulate the data to get the answer they want. That is a big no-no in science. MIT should know that is what he is doing? I demand and investigation (sarc).
I shudder to think what else Lindzen and Watts have talked about by email that they are not willing to share. Perhaps someone now has grounds to FOI Lindzen’s email?
Which simply means you don’t understand what reviewers are looking for when they review – dhogoza
Ah yes, peer review too is beyond the capability of mere mortals to comprehend.
Doesn’t look like peer reviewers (at least in some segments of climate science) are looking at or for anything.
The methods are included in the paper and are supposed to be sufficient for someone skilled in the art (sorry – patentspeak) to replicate the results – Deech56
Obviously you did not read the testimony of Phil Jones. Equally surprising as his testimony is the utter lack of interest by other climate scientists in verifying any of his results through replication.
Scientists do not typically ask other scientists to share code. That is not how we independently verify the validity of others’ research. We do it ourselves, and if both party’s got it right then we get the same answer. – MapleLeaf
Please don’t try and pass off such nonsense on the public. If I say the answer is “5” but my raw data is “1+3=?”, my results are not verified if you come along and say what it the answer to “1+4=?” and answer “5” also. That is nonsense.
[DC: Knowledgeable folks are patiently trying to explain the facts, yet you continue to wilfully misunderstand.
Jones stated in his testimony:
a) Raw station data was being released, subject to providers’ agreement (most but not all have agreed). The station data is also available from the original sources.
b) ML is absolutely right that it is the methodology as published, not the particular computer code, that is of interest to scientists.
c) Of course, reviewers are “looking”. I will admit in the case of papers by the likes of McLean et al, or Lindzen and Choi, it does appear that the system is being gamed – by the contrarians.
Now, I’m afraid I must accord myself the last word on the subject of release of material by Jones. Thanks!
BTW: Testimony at UK Parliament S&T Committee is here. ]
Paul S, a reviewer does not have the time to get the raw data and recreate the results of a submitted manuscript; the author is expected to be able to do calculations. In the case of the temperature data, verification of the CRU temp means getting similar results from NASA-GISS or satellites. It might mean using the publicly-available data and an algorithm of one’s choice to recreate the global temperature record.
What scientists are interested in is the big picture and the big questions, like what are the trends in the temperature record.
If you want to know what is involved in peer review, please ask. Some of us have had the experience of being the reviewer and the reviewee. Peer review is not hard to understand, but the general public may not be privy to the “Instructions for Reviewers” information.
Weilding his own ignorance as evidence is PaulS’s usual schtick. He’s a troll stuck on a replay loop no matter how many times anything is explained to him. He’s best flushed.
[DC: I err on the side of leniency, no doubt. But my patience is not infinite. ]
Anyhow DC, if you go to the last page of results at you google link, you’ll find that there’s actually about 600 results. This seems to be a regular glitch with google.
[DC: Bing is a lot worse, although Google may be exaggerating (see the Lawrence Solomon/ClimateGate post).
But note I can get 1000 results (the maximum that Google will actually return), if I allow duplicates. Here’s 991-1000. ]
What scientists are interested in is the big picture and the big questions, like what are the trends in the temperature record. – Deech56
And it seems like some scientists are only interested in studies that confirm a warming trend. Hence, no desire to see raw data, no curiosity about the methodology, and certainly no attempt at replication.
[DC: You’re not listening. The methodology is published, the vast majority of raw data is available from other sources (and more and more being made available all the time), and the work has been replicated, only not in the trivial sense you suggest. ]
Considering the importance given to temperature reconstructions done by Phil Jones, I find it incredulous that not once, ever, did a single scientist worldwide ask for any data relating to his work.
[DC: You are confused. The previous discussion was about the instrumental temperature record, and now you are raising reconstructions for some reason. ]
Can you explain how or why such important work would never be examined or replicated?
[DC: Again your premise is mistaken. Enough is enough. I’m afraid you’ll have to leave us alone now, since you are being quite redundant. Thanks!]
“Can you explain how or why such important work would never be examined or replicated?”
Nobody is saying that replication isn’t important, but scientific replication means finding the same result using independent means, not checking to see whether the calculations were done correctly. This is what happens after peer review. This is a very important part of the scientific method.
If you are saying that the CRU temperature record cannot be trusted, then all you have to do is ask yourself whether anyone else shows increasing temperatures over the last few decades (or longer). Answer – GISS and the satellite records confirm the results; therefore, the CRU temperature has been independently confirmed.
The problem with exact confirmation is that researchers (“auditors”?) need to go back to the data originators to get all of the national met data. If the data are not CRU’s to give they cannot give it. In fact, you can tell the reliability of sources by whether they concede this point.
Reading that transcript of the S&T committee testimony makes my blood boil.
Nigel Lawson getting yet another platform to state that the “hockey stick” is “largely fraudulent”.
[DC: Lawson’s a real piece of work. I do wonder how the GWPF ever got educational and research charitable status. The website is just an upscale Climate Depot, with a few peerages among the board for extra pretentiousness. ]
Phil Jones said, “most scientists don’t want to deal with raw station data, they want to deal with the derived product.”
Could it be similar to how the Viking Orbiter beamed down “raw” data in 1976 and then it was “cleaned up” so that scientists could then study the Martian features?
If that is what he is saying then it makes perfect sense! I have put an example up to show this. Hopefully some skeptics will understand.
It could be said that what I have made is the “derived product”. Now scientists can analyze it as much as they want to without the time consuming task of having to clean it up. The important data is there for all to view.
In UK, almost anything can claim that status, and GWPF is too new to have filed any records. here’s teh relevant entry in the database, which is worth reading.
I’m told that in fact, the government does check these things as results come in.
Also, let us recall:
Monckton: SPPI, and I think he learned the technqiues from Ferguson.
GWPF seems to have one real employee, Peiser, a selection from the usual menu of advisors, and unknown funding, although I’m told by a UK friend the Board is “really old, really Tory.”
Finally, it’s all in the family. Lawson’s son is married to Monckton’s sister…
Speaking as someone from the UK, I am pretty concerned about the activities of Lawson’s Global Warming Policy Foundation (GWPF). I have noticed that when Lawson speaks on behalf of the GWPF he is often quick to state that teh GWPF exists to lobby on GW policy, not the science (usually before going on to regurgitate some denialist guff). I wonder if the claim they are focussed on policy is the basis of their charitable status? I can’t see how Lawson can appear in front of the S&T committee in his role as Chair of a charity and accuse scientists of fraud. If there any lawyers out there the GWFP charitable status might be worth investigating.
I, too, am unsure of how the GWPF got charitable status – supposedly for Education/Training, by providing advocacy / advice / information, & sponsoring or undertaking research. Their aim :
‘TO ADVANCE THE PUBLIC UNDERSTANDING OF GLOBAL WARMING AND OF ITS POSSIBLE CONSEQUENCES, AND ALSO OF THE MEASURES TAKEN OR PROPOSED TO BE TAKEN IN RESPONSE TO IT, INCLUDING BY MEANS OF THE DISSEMINATION OF THE RESULTS OF THE STUDY OF, AND RESEARCH INTO
A THE SCIENCES RELEVANT TO GLOBAL WARMING
B ITS IMPACT UPON THE ENVIRONMENT, ECONOMIES AND SOCIETY
C AND THE ABOVE-MENTIONED MEASURES’
How does that fit in with what a charity is NOT supposed to do, i.e.
“18 No organisation can be charitable if:
it is created for political aims.
19. The following are examples of organisations or aims which are often assumed to be charitable, but in fact are not:
the promotion of political or propagandist aims, or the promotion of a particular point of view…”
That sounds exactly like what the GWPF was deliberately set up to do.
Perhaps further investigation is necessary…
[DC: It will be an uphill battle, although I must say if the GWPF passes, then the bar is even lower than I thought. Still, the list of charitable status think tanks is shocking. Here in Canada, they include such climate contrarian disinformation outfits as the Fraser Institute and the Frontier Centre for Public Policy, both of which are small-c conservative and have links to Canadian PM Stephen Harper’s Conservative Party. In my opinion, this needs to become a political issue.
I am planning a post on the GWPF, and I’ll be noting parallels with the Canadian situation, where the charitable Calgary Foundation funds the charitable Frontier Centre (and may still be funding Friends of Science through the back door).
When Lawson claims that he receives no funding from energy companies or prominent individuals in that business, it should be well understood he means no *direct* contributions. Sources of funding have simply been put through an extra layer via other charitable foundations. And he can’t even make that statement with regard to all direct contributions, since anyone can make an anonymous donation via PayPal (although they might lose the tax deduction that way).
For now, I’m also interested in the possibility the the GWPF may have coached Roger Harrabin of the BBC (which is the reason I brought up GWPF). Whatever the case, Harrabin should come clean about this – I have heard an audio interview where he seems to suggest that these were legitimate questions, which is astounding to me. Another future post – sigh – will it ever end? ]
On the surface, their aim reads fine by what the charity commission says. The problem is of course that they are carrying out propaganda rather than sciences, the issue I think would be to demonstrate that, and it is easiest to demonstrate with their track record. So we need to leave them time to build one.
[DC: I fail to see how regurgitating contrarian nespaper opinion pieces can possibly be confused with “presenting research”.
And by the way, unlike Climate Depot, they often display the whole article. That is likely a violation of copyright law, if they have not obtained permission to do so. ]
I was wondering about this while reading your piece, DC: did you manage to contact Harrabin and ask him about the background of the question?
[DC: Not yet. I need to check exactly what he said about the piece before I contact him. That audio is here.
Listening from 3:55 on, Harrabin seems to misinterpret the nuances of Jones’s answers, and refers to Jones’s confirmation of “substantial areas of uncertainty”. Harrabin’s comments about “two previous warming periods” and MWP support the idea that he doesn’t understand that his questions are not really valid, but merely rhetorical contrarian talking points disguised as questions.
I want to emphasize that I don’t think Harrabin is biased to the contrarian side, let alone in bad faith. But he appears to be misguided and confused about the science. In other words, he has more in common with Fred Pierce than with David Rose.
In a recent edition of “Harrabin’s notes”, entitled “Climate Armistice”, Harrabin had this to say:
Yikes! At the very least, he needs to name these supposedly well-informed bloggers. (And, yes, all of the above should be considered notes for a future post). ]
In your opinion, why are the majority of people, especially those with higher education, either ignorant of the climate change issue. Or more importantly not prepared to accept the facts and actually adapt to future changes?
[DC: That’s a big question. But I think the above discussion shows part of the answer – powerful interests have mounted PR disinformation that has bamboozled most of the media.
The issues are difficult enough for those with less scientific knowledge – add that sort of confusion and spin and the result is inevitable. ]
In the US, among 501(c)3 (nonprofits) are the George Marshall Institute and Singer’s SEPP.
As for GWPF, my friend says more of the Board:
“really old, really Tory, pretty wealthy.” If someone is in UK, run down the Board and see.
I would conjecture: Lawson and his friends are the likely funders, plus whatever loose change they can pick up.
BTW, DC: do you have a referenceable source for the charitable status of Fraser, FoS, etc?
[DC: Fraser Institute donation page is here. Frontier Centre’s is here and states that “Donation over $100 are tax-receiptable.” I’d have to dig deeper for the exact designations.
FoS does not have charitable status which is why the support by the Calgary Foundation was problematic, although apparently no complaint was ever upheld. ]
Deep – a couple of points:
* google’s “number of hits” numbers for complex queries like that are notoriously wrong – click through a few pages and you may find only a few hundred, rather than millions.
[DC: It maxes out at 1000 as I explained above. But you may well be right. For one thing the query “Phil Jones” only returns 749,000 hits. ]
* I’m sure Monckton was quoted from Copenhagen saying “no statistically significant warming in 15 years” (slight modification of the since 1995 date).
[DC: Now that you mention it, I recall seeing another comment to that effect, perhaps at John Quiggan’s I think. But Harrabin used the “1995” version of the talking point, so I think maybe he didn’t get it from Monckton. Sure is loud in the echo chamber though …]
Just in case anyone missed it, a short Monbiot v Delingpole debate from yesterday which ends up with Delingpole looking annoyed – not surprising, given the way Monbiot shows him up :
On tax-free in Canada…
CRA search page tells me hits for FraserInsitute.
Pingback: Grasping at Straws « Wott's Up With That?
I was not aware of this site. “WoUWT” seems to fill an important niche.
Out of idle curiosity, does anyone know what “quite close” is? 90%? 93%?
If not, no worries. When I have a bit of time I’ll track down the numbers and run them to find out.
[DC: I’ve checked Jones’s calculations using HadCRUT annual temperature anomalies for 1995-2009. The slope was 0.115C per decade, with a standard error of 0.059C. The corresponding p-value is 0.072, i.e. 92.8% confidence. ]
Thank you, DC.
As many here probably know there’s nothing magical about 0.05 that suddenly turns a result from “not a real effect” to “a real effect”. The 0.05 as a cut-off is a relic of tables in books before computers calculated p-values automatically. It has become so ingrained we still use 0.05 as a cut-off even when there isn’t any good reason to do so.
Some statisticians argue doing away with p-values, or reporting other values instead/along with, or just reporting the p-value rather than using a fixed level of significance etc–there are some vigorous debates on various stat threads, and in the literature too that are fun to read.
But how do you summarize all that for someone (like a reporter) whose only knowledge of stats is the 0.05 value = truth, if they even have that? (rhetorical question, btw).
Anyway, I’d certainly regard 0.072 as significant, but I can understand why Dr. Jones didn’t say this himself–it would have been easier to explain it the way he did rather than past history and current debates about p-values.
Sorry if this is a bit O/T.
Dan, you make a good point; 0.05 is somewhat arbitrary, but I think we should be cautious. One of the strongest arguments against the “cooling” meme is that the time period chosen is too short for the trend to be significant.
It might be better to note that the 1995 starting date is deliberately picked; it is the earliest starting time that does not give a significant trend at p=0.05 (assuming that’s true – I am sure it is, but haven’t done the calculations).
[DC: Yes, it’s so for HadCRUT (for this particular test). ]
It is important to note that 20-30 years of data are needed to know whether a trend is statistically significant. This is at least a consistent argument. In my day job I help design studies in which effects of a given treatment are studied. We would never deliberately underpower a study; by the same token, we should be cautious when choosing how many time points we use to make a conclusion about a trend.
[DC: At 25 years one gets 99% confidence. ]
1995 is the earliest year when the statistical significance of the trend from that year to 2009 safely fails.
There must of course be some such year, so by this Motl’s logic there are no trends at all, anywhere.
the answer, “Yes, there’s been no statistically significant trend for 15 years”, shows that even if there’s anything such as “global warming”, it’s not really worth talking about.
But since 15 is just n, where n is the largest value for which there’s no statistically significant trend, and there’s some n for every phenomenon, nothing is really worth talking about.
Motl is a fascinating example of how ideology produces irrationality and extreme ideology produces extreme irrationality.
Note that you could split up the 1975-2009 period into 1975-1984, 1985-1994, and 1995-2009. And pop goes the high significance for each sub-period. No global warming for 1975-2009. Divide and conquer!
[DC: The period 1997-2009, can be subdivided into two cooling periods.. Climate science is so much fun! ]
I think Monckton made the 95% claim on video while talking to young protesters the day after his talk was interrupted. I think the link was posted on Desmogblog or Climate Progress.
But since 15 is just n, where n is the largest value for which there’s no statistically significant trend, and there’s some n for every phenomenon, nothing is really worth talking about.
A nice, concise rebuttal.
Tamino has a good post about trends, here.
I don’t know nearly enough about statistics to be able to make this sort of argument myself, but it seems to me that if you want to see whether there is a genuine trend in a set of data, you have to find out whether there is a particular subset of data for which the trend varies very little, regardless of where in the data series you take that subset from. For example, if every 10-years’ worth of data produces roughly the same trend, regardless of where the 10 years starts and ends, then you could be fairly confident that you’ve identified a trend. It’s also worth taking into account the nature of the data – if you know there is a periodical rise and fall in the data then you would make sure that you don’t look for trends in periods that start at a high point and end at a low point, or vice versa. You would probably want to look at periods which match the regular fluctuations, or multiples thereof. That’s what I tried to do here:
Note: just to make sure there is no confusion, there is no magic about n=15, or n=30, or any particular n. I.e., this is unrelated to the n=30 that often pops up in statistics.
What matters is, more-or-less, is the ratio of yearly noise to yearly trend, so that a larger ratio requires more years to detect a trend with increasingly significant results. With enough noise (say 5X larger than yearly trend), statistically, one *will* have multiple-year sequences of counter-trend periods.
For example, in comparison with temperature trends, it takes less years to see significant trends in CO2 concentrations.
As illustrations of all this, try graph, including animated one, or slope graph: 5,10,15,30-years.
That graphs the Excel SLOPEs (linear regressions) of N-yer periods by ending-year, so that you can see, year-by year, the trends to that year, looking backward N years. This converts the difficult perceptual problem of comparing diagonal slopes to a much easier one of comparing heights. it is clear that even 10-year slopes haven’t been below zero for a while, that 5-year slopes jiggle around, and 30-year ones don’t very fast. Of course, that chart didn’t show significance levels.
[DC: I’ve reposted this in the new Open Thread # 2. ]
Pingback: Flogging the Scientists « Climate Denial Crock of the Week
“California Attorney General Jerry Brown
What is his dream?
The interests of the environmental movement in acquiring more power and influence are reasonably clear. So too are the interests of bureaucrats for whom control of CO2 is a dream-come-true.
After all, CO2 is a product of breathing itself. Politicians can see the possibility of taxation ….”
— Feb. 2008, Lindzen
Pingback: Phil Jones - Warming Since 1995 is now Statistically Significant
DC – about 11 minutes into this recent interview Monckton claims the credit as the originator of the 2010 question put to Phil Jones:
(warming, fawning interviewer may provoke nausea)
Personally, I don’t believe him (why wait nearly 2 years, and why believe anything Monckton says in any case?). Of course, Monckton omits to mention that subsequently warming since 1995 was confirmed as statistically significant by Phil Jones.
The meme originated with Lindzen, Watts and Motl, but how it got to Harrabin is hard to say, as he only referred to “sceptics”. I wish I had asked him at the time – he may have forgotten by now.
DC, the usual suspects regularly send emails to legislators and the press. I’ve been surprised by the diversity contained in a single list of recipients sent the latest tosh, including the British Prime Minister. Fling muck widely and often, and something will stick to someone’s clothing.