NOTE: See the UPDATE at the end of this post.
I didn’t expect Judith Curry to embarrass herself more than she did with her fawning over Murry Salby’s folly. But she’s topped (perhaps I should say “bottomed”) herself by a huge margin.
Anthony Watts was so excited he actually suspended his blog hiatus to report the story. He quotes the GWPF that “BEST Confirms Global Temperature Standstill,” and cites a story in the Daily Mail reporting that Judith Curry (a member of the Berkeley team) has roundly criticized Richard Muller (leader of the Berkeley team), on much the same basis. He quotes Curry herself, from the article:
As for the graph disseminated to the media, she said: ‘This is “hide the decline” stuff. Our data show the pause, just as the other sets of data do. Muller is hiding the decline.
With this statement, Curry reveals how little she understands the data created by the team of which she is a member, let alone the “other sets of data.” Which we might have anticipated, given that from what we’ve seen, she made little or no contribution to the actual Berkeley analysis.
Judging by her comments, it’s a good thing for the Berkeley team that she didn’t. The Daily Mail quotes her thus:
‘There is no scientific basis for saying that warming hasn’t stopped,’ she said. ‘To say that there is detracts from the credibility of the data, which is very unfortunate.’
That’s a pretty strong statement, Judith. Do you have a scientific basis to back it up? Have you actually analyzed the data from your own team? Do you even know?
Now that the Daily Mail and Anthony Watts have not only let the cat out of the bag but put it on display to make a “scandal” out of it, Curry seems to want to distance herself from her criticism. On her own blog she says:
I told Rose that I was puzzled my Muller’s statements, particularly about “end of skepticism” and also “We see no evidence of global warming slowing down.”When asked specifically about the graph that apparently uses a 10 year running mean and ends in 2006, we discussed “hide the decline,” but I honestly can’t recall if Rose or I said it first. I agree that the way the data is presented in the graph “hides the decline.” There is NO comparison of this situation to Climategate. Muller et al. have been very transparent in their methods and in making their data publicly available, which is highly commendable.
My most important statement IMO is this: ‘To say that there is detracts from the credibility of the data, which is very unfortunate.’ My main point was that this is a very good data set, the best we currently have available for land surface temperatures. To me, this should have been the big story: a new comprehensive data set, put together by a team of physicists and statisticians with private funds. Showing preliminary results is of course fine, but overselling them at this point was a mistake IMO.
It seems to me that there are only two possibilities. Either Judith Curry hasn’t bothered to analyze the data from her own team — or she’s not competent to.
Just looking at the data isn’t enough. It’s all too easy to view some graph and get the wrong impression — in fact that what groups like GWPF exploit. Here, for instance, is the graph featured by GWPF:
The impression it gives, that global warming has stopped, is wrong. It’s the kind of easy misinterpretation based on the absence of both context and analysis which is constantly exploited by fake skeptics. I expect GWPF to disdain both context and analysis. But Judith Curry is supposed to be an actual scientist. I’m skeptical.
Since Judith Curry seems unwilling or unable to do the analysis, we’ll just have to do it for her. Let’s start with the data used by GWPF. Here it is with a trend line:
Please note that it starts in January 2001 and ends in May 2010, so it covers a tiny bit less than a decade! That’s a great big “red flag.”
The slope of the trend line (from ordinary least squares) is only 0.03 deg.C/decade. It’s fair to call that “flat.” But the standard error from that calculation — even if we use a white-noise model — is 0.13 deg.C/decade. So the real value could be as high as 0.29 deg.C/decade — which, curiously enough, is a tiny bit higher than the trend rate in the Berkeley data from 1975 to the present — a period for which you can only deny warming if you’re both blind and insane. And yes, it could easily be an even higher slope since we’ve used a white-noise model, which underestimates the uncertainty. Hey Judith: there’s a scientific basis for you.
But that’s not all! The Berkeley data also include uncertainty levels for each monthly value. Let’s graph the uncertainties:
Whoa!!! What’s that huge spike at the end? It rises faster than a “hockey stick”!
The huge spike is the huge uncertainty level of the final two data points — April and May 2010. While the other data have uncertainty levels around 0.1 deg.C, those two months have uncertainty levels of 2.8 and 2.9 deg.C. Which makes them, plainly, unreliable.
And notice that one of those too-uncertain-to-be-useful data points is the deep “dip” at the end, the April 2010 value which is about 2 deg.C below the trend line. It’s called an “outlier,” in fact it’s an outlier of the first magnitude, a really big super-giant “this ain’t right” staring you in the face. And since it’s a super-giant dip it’s gonna skew the trend estimate low. The simplest correct thing to do is to omit those two data points.
So let’s do it again, but end at March 2010 rather than include the uber-fuzzy April and May 2010 data. Here’s the trend line from 2001:
Now it’s noticeably upward. The estimated slope is 0.14 deg.C/decade — more than four times as large, just from removing two errant data points. Its standard error is 0.11 degC/decade, so the real trend rate could be as high as 0.36 deg.C/decade, quite a bit larger than the average rate since 1975. And yes, it could easily be an even higher slope since we’ve used a white-noise model, which underestimates the uncertainty. Hey Judith: there’s another scientific basis for you.
To show just how sensitive the results are to such a short time span — and how misleading it can be to make pronouncements on that basis — let’s consider a time span just one year longer, starting in 2000 rather than 2001. That gives this:
Now the trend rate is up to 0.27 deg.C/decade. That’s just about the same as the trend rate from 1975! Hey Judith: there’s another scientific basis for you.
As a last bit of analysis, let’s omit the outlier data, then compute the trend for each starting year to the present, for all start years from 1975 to 2005. Let’s also estimate the confidence intervals using a more realistic error model. What would that show? This:
The red dashed line shows the trend rate from 1975. Note that not one single start year gives an estimate which contradicts that rate. That is evidence — damn strong evidence in fact — that the underlying trend rate has not changed since 1975. Hey Judith: there’s a REAL scientific basis for you.
Here’s the kicker: the uncertainty in those trend rates is probably higher, perhaps by a substantial amount, because that graph is based on an AR(1) model for the noise. Using an ARMA(1,1) model instead gives this:
That shows just how mistaken, how foolish, how downright boneheaded it is to say that “There is no scientific basis for saying that warming hasn’t stopped.”
Judith Curry’s statement is exactly the kind of ill-thought-out or not-at-all-thought-out rambling which is an embarrassment to her, and an embarrassment to science itself. To spew this kind of absolute nonsense is shameful. Judith Curry, you should be ashamed of yourself.
Richard Muller, the target of Curry’s ire, also managed to embarrass himself:
However, he admitted it was true that the BEST data suggested that world temperatures have not risen for about 13 years. But in his view, this might not be ‘statistically significant’, although, he added, it was equally possible that it was – a statement which left other scientists mystified.
Of course we’re mystified. Didn’t you do the analysis? Don’t you know? Honestly, what kind of insane world do we live in when we get this level of discussion about one of the most crucial issues of our time, from people who are supposed to be scientists?
In my opinion, the WORST aspect of this is that Judith Curry states unequivocally that “Our data show the pause,” which utterly ignores the extreme level of uncertainty in temperature trends over periods as short as 10 years or less, after she has so often indulged in self-righteous posturing on her own blog about the “uncertainty monster.” It looks like the “uncertainty monster” decided to turn around and bite her on the ass.
Regarding Curry’s involvement with the press release, and the submission of papers, the Daily Mail reports that “Prof. Muller failed to consult her before deciding to put them on the internet earlier this month, when the peer review process had barely started, and to issue a detailed press release at the same time.” She stated, “It would have been smart to consult me.” On her blog she says:
My continued collaboration on this project will be discussed this week with Muller and Rohde.
It seems to me that what would really be smart is to find someone else to be on the Berkeley team.
UPDATE:
Judith Curry protests that she was misrepresented by the article in the Daily Mail, and several readers have mentioned that David Rose, the author of the article, is just the man to do such a thing. It’s easy to believe that she was indeed the victim of his malfeasance.
But even after reading this post, she still hasn’t disavowed the statement “There is no scientific basis for saying that warming hasn’t stopped.” In fact she commented on her own blog saying, “There has been a lag/slowdown/whatever you want to call it in the rate of temperature increase since 1998.” Question for Curry: What’s your scientific basis for this claim?
Open Mind 
In the sense that some perfectly solid science is being blatantly misrepresented by shameless hacks who lack the required knowledge to contribute anything of worth; yes. Yes, it is.
Woodfortrees now includes the BEST data so people can easily graph it themselves. Plotting a 10 year mean either since 1975 or since the start of the record gives a good idea of what’s going on. It’s pretty clear the 2000s didn’t warm at the same rate as say the 1990s but also that there’s nothing unusual or unexpected about that. It’s very difficult to argue for either a pause or cooling looking at the data in that way.
Plus the laws of physics didn’t go on hiatus within the last 10 years. Whatever people’s pet theory for a cause of cooling unless it’s as long-lived as C02 is then we’ll simply get a jump in temperature whenever it stops.
Oh come on:
http://www.woodfortrees.org/plot/best/mean:120/from:2005
It’s obvious.
:-)
Nice report. Thanks for the analysis. More interesting than the debate over data, is the biased human thinking that Curry displays.
This kind of insanity is explained by inability to face the enormity of the sin. Human self-destruction has been actively promoted – and we are carrying it out, and nobody is stopping it. Our natural tendencies of greed and willful stupidity have been deftly amplified by carbon culture.
This is so difficult to grasp that people retreat to lesser thinking – using all the denial and logical bias that one can possibly summon.
I love your comment. At least some people have the faintest clue as to what is going on. These are not isolated phenomena from the poisoning of the water on the whole freaking planet. Poisoning the soil and wasting the accumulation of top soil of many centuries for a few bucks more.
I’m too old to really care. It just seems totally ridiculous that somebody somewhere believes very very insanely that this is not going to screw up his descendents too. What are the fools planning on, flying off to flippin Mars?
I must be missing something, because according to my analysis the data is clearly flat for the last decade. The analysis above is complete bull. He only states the result of ADDING the standard error. Isn’t it statistically equally likely that subtracting the error gives the correct values? If so, wouldn’t those results show a DECREASE? If you are going to use the “error” to evaluate what the date COULD be, you need to look at BOTH possibilities.
You (all of you) must live in a different world than I do.
[Response: On that we agree]
I have no concern about the future of either our planet or our species. The technology that we have developed (and the energy needed to power it) has created a world that today is better than at any time in the past, and I have no doubt that we will continue the trend with future developments. I have faith in mankind…
Oh, that one is a classic!
“Isn’t it statistically equally likely that subtracting the error gives the correct values?”
Sure, Ringworlder. But “equally likely” is not a basis to say that warming has stopped–particularly when the past record shows that there have been similar “pauses” in warming, and when the underlying physics haven’t changed.
In my (spherical) world, at least.
I see that the ‘Mail on Sunday’ piece was written by David Rose, so that tells us how seriously we should take it.
Its fun to note that BEST’s FAQ takes on this question directly. JC must have been absent that day.
It’s also an important point that BEST does not include the summer of 2010, which I recall was a toasty one.
Curry is busy over at her blog trying to correct/ spin (delete as applicable) the Mail on Sunday article.
After referencing a “paper” posted on the EIKE website (EIKE=CFACT) on that same thread, no need to think long about what is most applicable.
“Quick, move the goalposts!”
As odious as Rose usually is, I’d be curious to hear his comments on his interpretation of the exchange between himself and Curry. I’d especially like to know whether he (or she, for that matter) thought to record the conversation for fact-checking and for corroboration.
And Curry thinking that speaking with Rose at all is a good idea? No, Judith, no. Just no. It’s a very telling reflection on your capacity for objective, reasoned thought.
Using wood for trees, I posted the 120 month graph on the thread at Climate Etc. It shows an upward trend. Nobody said a thing.
The last calendar decade in the data, 2000 to 2009, shows a stronger upward trend. I also posted it.
Tamino,
You say “So the real value could be as high as 0.29 deg.C/decade ” and “so the real trend rate could be as high as 0.36 deg.C/decade”. To be fair, could those rates not also be -0.23 and -0.08 respectively, IOW why assume that the error only works one way? (If I’ve misunderstood, please put me straight.)
[Response: The final graph in the post summarizes the situation effectively.]
Yes, I can see that, but in the text I quoted, you mentioned only the “plus” possibility. Some might see this as a “warming bias”. ;)
[Response: Spoken like a “true skeptic” :) But the issue is Curry’s claim that “There is no scientific basis for saying that warming hasn’t stopped.” Fact is, there’s no scientific basis for her claim.]
Well, I hope I am, unlike the pseudosceptics that infest much of the climate blogosphere.
Of course, Curry’s claim is just plain stupid and she has the knowledge to know better. (David Rose, the GWPF, and Watts we know are strangers to the truth.)
This really just boils down to “x years is too short a period for noisy data”, doesn’t it? I doubt that the fantasists will ever stop trying to use that trick to mislead.
Of course it could be. But if you did this analysis back in 1995 or in many other years back then, you would exactly see the same thing. But the warming didn’t stop there, and it did not stop at any of the other point in the last 40 years. So, if you look at a short time period of just a few years, the uncertainty is always higher than the underlying trend. You can therefore *always* say, that the trend in this small time span *could* be negative, even if you are in a strong warming.
But would that be reasonable? Obviously not. As long as the temperatures are within the uncertainty range of the warming trend, there is simply no reason to say, that the warming trend has stopped.
This logic, that it *could* have stopped, is just an irrational argument. Look at this analogy: You are in a plane, the only pilot got a heart attack and is unconscious and the landing gear malfunctions. Well, you *could* say, that, despite this, a passenger *could* land the plane so safely, that everyone survives this. But would that be a reasonable expectation, just because it *could* be? Would this be a reason, that you needn’t to worry anymore about the landing? No! Analogously, we are in a strong warming trend over the past 40 years and if that continues, we will get very likely in serious trouble. So, just because the uncertainty over a small period of time is so high, that it allows you to say, that it *could* be cooling and therefore we don’t have to worry about global warming anymore, this would be as reasonable to believe that you don’t have to worry about the landing of the plane in the example above.
I cannot keep up with all of it, but the legitimate papers I know of making a prediction of cooling are failing – especially using GisTemp or NOAA.
The ones that come to mind are Keenleyside et al and Tsonis-Swanson.
In the other direction, Smith et al, looks reasonably accurate to date and is still in play.
So would this be a good time to remind people of John McLean’s prediction of global cooling for 2011?
(Yeah, that peer-reviewed paper. Strangely, the website doesn’t mention the tenor of the post-publication peer review.)
Thank you, Lotharsson–that’s definitely appropriate as an update for my “When did Global Warming Stop” article!
@TrueSceptic: “IOW why assume that the error only works one way?”
What all these graphs and numbers show (and there are many other ways to come to the same conclusion) is that we can’t really conclude anything useful based on the last ten years (or any decade’s worth) of data, by itself, because there is too much variability compared with the presumed trend. Yes, the longer term trend may be continuing. Yes, the land temperature increases may have leveled off. Yes, the increases may have accelerated. But we don’t have any statistically valid basis for concluding any of those things based on this data. So when Curry says “There is no scientific basis for saying that warming hasn’t stopped”, she could just as well have said “There is no statistical basis for saying that warming *has* stopped”, based just on these last 10 years or of of data. She’d be equally right (if you call it “right”) either way. It’s a purely useless statement, scientifically speaking.
To Judith’s defence, the honourable David Rose has quite a track record in misrepresenting scientists in favour of spreading misinformation. That doesn’t give her free pass to spout utterly silly things though…
How does that defend her?
Tamino, I see that the article in the Daily Mail was written by David Rose. I would note that Deltoid has a category devoted to David Rose called “Rosegate” which after reading should lead you to conclude that when David Rose “quotes” someone you have no idea what they really said.
Actually, you could reasonably assume that they said something more likely to be the opposite of Rose’s “quotes”.
Ah, but she didn’t, for the most part!
Rose just makes up the context.
TrueSceptic:
The point is that Curry’s position is based on an assumption that either there’s little uncertainty or the uncertainty always points in the cooling direction. She’s made this clear in a multitude of posts on “uncertainty in climate science”.
So, putting her position on its head as Tamino does effectively skewers the stupidity of her assumptions.
Just IMO …
If I was still a sceptic I’d seriously doubt the competence of the company I was keeping. Fortunately for me I hit that point years ago, so I’m spared this sort of train-wreck (pseudo?) science.
The last graph is worth a thousand words.
Good rapid response Tamino. Unfortunately the Hate Mail’s tactics are working. They appear to be following Anthony Watts in juxtaposing GWPF with Muller.
BBC Radio 4 reported all its main points this morning. A few days ago the Guardian ran an article about misleading information in the Mail but I did not notice that Radio 4 gave it a comparable mention.
Will Radio 4 listeners ever be informed about Tamino’s response>
As Tamino and others regularly point out, a 10 year span is too short to report confidently on any change in the global temperature trend. Given that 7 of the 8 globally warmest years have occurred since 1998, proclaiming a decline or even a halt seems downright idiotic. This simply shows that many scientists are ill-equipped to provide off-the-cuff statements to the mainstream press, let alone media with ideologically pointed agendas.
Yet, to refine our climate projections, we must still look deeper than just debunking the deniers.
E.g., given that the net radiative balance at the top of the atmosphere remains negative, which certainly indicates continued warming, Trenberth’s studies suggest deep ocean uptake of most of the recent heating. Meanwhile, Hansen had suggested that recent aerosolized sulphate increases in our troposphere (perhaps largely due to China’s proliferating, unscrubbed coal-fired power plants) has modulated more recent warming rates.
In any case, much work is ongoing to achieve consensus on the current hierarchy of second order forcings and feedbacks. Once achieved, we must quickly disseminate that new information to the public, hopefully in such a way as to minimize inevitable, denier-driven distortions.
It seems GWPF’s lap dog David Rose has been at it again. Curry has been played, it seems. At least, that is what she seems to imply on her blog.
Curry replied to my comments on her blog:
“Well, I have a rule about not talking to reporters on the phone, asking for submitted questions and I respond by email. Its a rule I extremely rarely break, and Rose caught me on the phone and I spoke with him. Back to enforcing my rule.
IMO Muller overplayed his hand in his media statements, why I am not sure.
I am getting a substantial number of press queries, I will do my best (lowercase) to set all this straight (via carefully crafted email replies).”
and
“At the moment, I’m feeling manipulated by both Rose and BEST. This is one reason I started a blog, to get my words out there and minimize my personal exposure to manipulation.”
I suggest Rose deserves more of a correction.
The April 2010 result seems so obviously anomalous that anyone’s first instinct should surely be to check what is going on with it. Fortunately if you look at the source data from BEST the 2.763 uncertainty figure is sitting there right next to the -1.035 anomaly, so not much work needed. Is there any reason to think that an anomaly of -1.035 popping up like that is justified in the observations? Did anything happen in that month that might have led to it? (The eruption of Eyjafjallajökull perhaps?)
The most likely answer is: Most stations did not report a temperature for the most two recent month, when they made their analysis. So April and May 2010 are based on a very small subset of temperature readings and huge parts of the earth are not covered for those month, hence the huge uncertainty. And if those small subset of stations is incidentally situated mostly in regions, which experienced a colder-than-average April, then you can get easily to the -1,035°C. There is nothing in particular wrong with this, as this low value is accounted for by the huge uncertainty.
But IMO it would have been better, to just limit the analysis up til March 2010, where you have most of the temperature data available and not incorporate two month with such a sparse amount of data. You can still add them, when the temperature data of those month is available. So, if you do some analysis with the BEST data, it is currently the best thing, to just omit the last two month and work only with the data from March 2010 on.
That makes sense, but why would they not just leave the final two months out?
TrueSceptic — Curry’s claim was one tailed, so Tamino focused on that tail. Tamino — This may be introductory text book stats, but I think I remember 95% of the distribution lying within 2 standard *deviations* of the mean. Here you are using standard errors. Am I getting it wrong, or is this a special circumstance in which the two can be used interchangeably?
[Response: For a single data point you should use the standard deviation, but for a model parameter you should use the standard error.]
Thanks for the response.
Note: See the UPDATE at the end of the post.
I sort of popped my cork when she said that.
I’ve learned so much form this blog, host and commenters. Thanks for getting up this morning doing great work.
“At the moment, I’m feeling manipulated by both Rose and BEST. This is one reason I started a blog, to get my words out there and minimize my personal exposure to manipulation.”
Ah, the victim card … she played it so well when called out on her shit the very first time she put it on parade at Real Climate, too.
Not to worry! Fuller and Mosher are providing Curry and the BEST team with public relations advice.
Glad you could replicate the effect of the April 2010 outlier; I’ve been banging on about it at WUWT:
http://www.woodfortrees.org/plot/best/from:2001/mean:12/plot/best/from:2001/trend/plot/best/from:2001/to:2010.2/trend
Eyjafjallajökull: Good call!
If the quote is accurate, it is weasel worded double negative statement–correct in a very narrow technical sense, but conveying an impression that is diametrically opposed for reality. For example, it is true that “there is no scientific basis for stating that the warming did not end 10 seconds ago.” Of course, there is no way there ever could be a scientific basis for stating such a thing regardless of whether it was true or false, but the statement is phrased as such a way as to give the impression that absence of evidence that warming has not just stopped equals evidence that it has.
I can easily imagine an unethical journalist manipulating a not-very-savvy scientist into “agreeing” with such a statement, much as Jones was manipulated into agreeing with a similarly misleading “no evidence of statistically significant warming over the last 15 years” statement.
Ah yes, pulling a “Rumsfeld”. We all know how *that* particular piece of wisdom panned out.
How about “The error bars on the last months of data from our study are too wide for us to advance ANY FIRM SCIENTIFIC CONCLUSION whatsoever, based on those last months.” ?
“I can easily imagine an unethical journalist manipulating a not-very-savvy scientist into “agreeing” with such a statement”
I’d like to believe this, but reading her commentary (and, in particular, lack of commentary) on the Mail’s article I’m becoming ever less convinced of Judith Curry’s good faith.
Starting in 2000 actually makes more sense than 2001, because that’s when the CMIP3 model projections start (before that it’s hindcasts).
BEST’s large data set seems to have created a problem in keeping the data up to date. What’s the point of a modern temperature record that is more than a year out of date? And why did they put out a data set that appears to contain incomplete data for April and May 2010?
I also suspect that while BEST broadly agrees with the other temperature records, that it may be out of sync at the month level, and possibly the annual. But perhaps errors engendered by their inclusion of masses of less reliable data, are cancelling out over time.
It’s also worth noting that Muller himself has been careful not to pronounce on AGW, only the validity of the global land temperature record.
I would think you should start in 1999 using that logic. If you’re calculating the trend from the end of the hindcast period the 1999->2000 change shouldn’t be omitted.
Of course, if changing end points by a single year makes a big difference that’s a good sign the result isn’t meaningful.
Tamino,
Did you already have a chance to look at Rohde’s methods? http://berkeleyearth.org/Resources/Berkeley_Earth_Averaging_Process
It seems that the authors manage to avoid various problems, for example those caused by gridding. However my statistical knowledge is not extensive enough to judge the validity of what they are doing.
That said, the Earth (or land) average temperature seems to be such a stable parameter that even the crudest averaging still gives the same result (warming!). And earlier authors were of course not unaware of gridding problems etc, so the whole thing is a rather academic. To quote RC: the result is just as unexpected as the Sun rising at dawn.
[Response: I’ve only given it a cursory examination, but the method seems both logical and sound. In fact I think it might be the “best” part of their effort. I need to study it more closely, but what I’ve seen so far impresses me.]
If you do, any chance you could write a post on how it works? I read the paper, but some of the math was beyond me. It’d be nice to see a post summarizing it to make the big picture a bit more understandable.
I cannot be bothered to go to Curry’s denier infested blog, so this might have already been addressed, but did she really say:
“As for the graph disseminated to the media, she said: ‘This is “hide the decline” stuff. Our data show the pause, just as the other sets of data do. Muller is hiding the decline.”
Because if she stands by that, it has some serious and bizarre implications. It suggests that she is accusing the BEST team of fudging the data (aka scientific conduct) without seeming to realize that she is part of BEST, so is implicating herself. She is also accusing some highly respected scientists (including a Nobel prize winner) that she worked with of scientific misconduct. But as muoncounter showed above, BEST very explicity preempted the expected cherry-pick that “the last X years of data have showed no warming”, so I am unsure how Curry can accuse her co-authors of hiding the decline.
I was unsure on which side Curry would fall when the expected furor over the BEST results came out. Well it seems that she is now siding with those who deny the reality that the planet is warming and those who routinely distort, misrepresent the data (i.e., Watts, Peiser and McIntyre, Monckton, Delingpole etc.). That is pretty pathetic, especially if she is making false accusations against her co-authors and while giving the GWPF and Watts and Delingpole a free pass. Those are the people who are bending over backwards to hide the incline and Judith seems to agree with them.
The stench of denial by Watts et al. is quite overpowering now, and McIntyre has been reduced to desperately cherry picking individual stations and spinning the data to reinforce his denial and to use as fodder for fellow deniers like Watts.
I must admit it is quite satisfying watching those who deny AGW feasting on each other. Curry must be delusional to think that Rohde and Muller et al. are going to take the unwarranted attacks on their integrity lying down.
Rose has posted on Curry’s blog, including this:
“…But I think your memory is at fault when you state that it was I who first used the phrase “hide the decline”. You did this, twice, in our first conversation, although it’s true I was the first to mention it in our second talk…”
Thanks Holly. Well then Curry might be advised to lawyer up at this point.
I was there when that popped up. Too funny. He does have a great reputation, right?
What is sad is that ordinarily it would be obvious who not to believe here (Rose), but given Curry’s recent track record, I honestly have no idea who is telling the truth, it may be that neither one of them is.
Either way, Curry continues to embarrass herself, show her disregard for elementary statistics and slander her scientific colleagues. She has chosen her tribe– Watts and GWPF.
Now Muller has some sense of how Mike Mann feels. I hope Muller also sees the irony after he only recently slandered Mann et al.?
ML:
Well, Curry’s also stated that she’s meeting/talking with them this week, and that her continued participation in the project is up for discussion.
She presents it as though *she* has put it on the table, i.e. she might quit because of their “unwarranted spinning” (my paraphrase) of the BEST results.
Methinks she might be in for a surprise, though, regarding who’s calling the shots regarding her continued participation …
1) I now regret having failed to schedule a visit to Santa Fe conference this week.
This session run by Curry will surely be fascinating:
“T-II: Observations (Judy Curry and Manvendra Dubey)
T-5: S. Wofsy (Harvard) HIAPER Pole to Pole Observations (HIPPO) of climatically important gases and aerosols 10:35-10:55
T-6: R. Muller (UC Berkeley) The Berkeley Earth Surface Temperature Land Results 10:55-11:15
T-7: R. Rohde (Berkeley Temp Project) A new estimate of the Earth land surface temperature 11:15-11:35
T-8: F. Singer (SEPP) Is the reported global surface warming of 1979 to 1997 real? 11:35-11:55
T-9: J. Xu (NOAA) Evaluation of temperature trends from multiple Radiosondes and Reanalysis products 11:55-12:15
Discussion 12:15-12:30”
Actually, if you look around the schedule, many other “interesting” speakers appear, but sadly, the Viscount is in a different session.
2) At BEST, we find the 4 papers, of which Judith Curry is the 2nd author (and hence, one would think very strongly involved) in all 4:
1) Berkeley Earth Temperature Averaging Process
Robert Rohde1, Judith Curry2, Donald Groom3,
Robert Jacobsen3,4, Richard A. Muller1,3,4, Saul Perlmutter3,4,
Arthur Rosenfeld3,4, Charlotte Wickham5, Jonathan Wurtele3,4
2) Influence of Urban Heating on the Global Temperature Land Average
Using Rural Sites Identified from MODIS Classifications
Charlotte Wickham1, Judith Curry2, Don Groom3, Robert Jacobsen3,4,
Richard Muller3,4, Saul Perlmutter3,4, Robert Rohde5, Arthur Rosenfeld3,
Jonathan Wurtele3,4
3) Earth Atmospheric Land Surface Temperature and Station Quality in the United States
Richard A. Muller1,2,3, Judith Curry4, Donald Groom2, Robert Jacobsen1,2, Saul Perlmutter1,2, Robert Rohde3, Arthur Rosenfeld1,2, Charlotte Wickham5, Jonathan Wurtele1,2
4) Decadal Variations in the Global Atmospheric Land Temperatures
Richard A. Muller1,2,3, Judith Curry4, Donald Groom2,
Robert Jacobsen1,2, Saul Perlmutter1,2, Robert Rohde3,
Arthur Rosenfeld1,2, Charlotte Wickham5, Jonathan Wurtele1,2
[Response: Notice the pattern: there’s a lead author, followed by all other members of the Berkeley team in alphabetical order. That’s why Curry is 2nd author on every paper.
My guess: her only contribution to the Berkeley effort was her name.]
Names: yes, a curious method. is that typically done elsewhere?
Bell Labs in my day was strictly alphabetical … which meant in practice that the lists generally included the people who’d really done serious work.
Actually the practice I am most familiar with is the first author being the one that pulled it all together, often a post-doc researcher. Then, the last position in the author list may be for the ‘senior author’, like the first author’s professor… but only if he had actual input (if only in the form of bright ideas, or reading it through and commenting before submission) into the paper.
Including an author name for someone who didn’t actually do anything borders on the fraudulent (although in many circles not at all uncommon from what I know).
…I would add — and many folks I know don’t appreciate this — that co-authorship requires a basis of trust; a bit like marriage ;-)
You’re going to be held to account for the whole paper, also those parts only your co-authors understand deeply. Remember that the value of multiple authorship is precisely in bringing together non-overlapping expertise…
Hi Tamino,
Thanks for the great post. For the last two plots how come they stop at 2005? Isn’t the outlier data only in 2010?
[Response: Those graphs show trend rates from the given time up through March 2010. It stops at 2005 because computing the data from, say, 2006 (or later) to the end is such a short time span that nobody would say it gives anything like a reliable trend. Except maybe … ]
Thanks.
Tamino, just to clarify, your last two graphs represent the confidence intervals for the underlying trend, not the actual trend correct? Because, correct me if I’m wrong, the actual temperature (IE the actual physical surface temperature of the earth) has much smaller confidence intervals. That is an important distinction to make and I am sure you have explained it elsewhere. But it is a difficult concept to grasp so I think it would be good to explain the distinction each time you make it.
Part of the problem is Curry’s language is not very clear. She’s probably talking about ‘actual’ temperature trends while you’re talking about the ‘underlying’ trend. But as a scientist discussing this important issue publicly, it is her responsibility to present the issue clearly.
[Response: Yes, the last two graphs show the trend estimate (from the given start time through March 2010) and its confidence interval, i.e., the confidence interval in the underlying trend.
As for Curry talking about “actual” trends — not so. What might look to you like an “actual” trend isn’t one, it’s just a consequence of the noise. Even white noise can give that impression, and with strong enough autocorrelation one can get hellacious apparent “trends” that are nothing but noise. Noise isn’t trend — by definition.]
Linear temperature trends over just 10 years have no scientific value IMO. It is a fact that world temperatures over the last 10 years have a linear trend near zero. The last time that occurred was in the middle of 1997. And then came 1998; by the end of that year the linear trend over 10 years was +0,32°C/10yr (mean of GISS/NCDC/HADCRUT/UAH/RSS).
I don’t know. Say there actually was a precipitous downward trend – say 2011 actually dropped to 1956, or many years before that, as mentioned above.
OK but by noise you do not mean simply measurement error, you also mean noise caused by non-GHG temperature fluctuations like ENSO and solar?
In other words, there is noise due to measurement error, but then there is also noise due to other climate variables which obscure the underlying trend?
[Response: Exactly correct (imagine, for instance, that the Berkeley data were *perfect* — that wouldn’t enable us to determine the trend with perfect precision). It’s also why it’s inadvisable to rely on the stated Berkeley uncertainties for a “weighted least squares” regression — they don’t include the noise due to natural variability.]
Hypothetically, we could have a statistically significant lack of warming since year X because of La Nina and solar minimum, but at the same time you would also be able say that the underlying trend may be between -.3/decade and +.3/decade.
I think Curry is clearly TRYING to say is that the trend since 2001 is not statistically significant from zero. But she needs to be much more clear that this does NOT mean it IS zero.. it could be zero, or it could be +.4C/decade.
[Response: I disagree. That’s certainly not what she said. I think she wants to assert that global warming has definitely ceased (according to her latest comment, in 1998 no less!). But so far she hasn’t given a “scientific basis” for such a claim. Not one peep.]
Ironically, Judith is ignoring her “uncertainty monster” by proclaiming the short term annual global temperature non-trend as significant in a scientific sense. Since annual global temperature data is moderately noisy due to natural fluctuations, trend significance can only be detected with high confidence after considering the physical basis. Either run a GCM through recent data or filter the natural fluctuations to focus only on inherent warming trends. I’m wondering what Judith will say on your trend paper Tamino :)
But the basic question still remains the same RE:
Did Dr. Curry ever read the draft copies of these papers at any time before the BEST PR of October 20, 1011?
I mean, if you are going to have your own name on said papers, don’t you think it would be useful to read said papers first?
It really doesn’t matter where her name appears on the list of authors.
So either she read them beforehand or she didn’t.
If she did read them, and had no edits, then she accepted them implicitly.
And to date, there is nothing to suggest that she had any editing input at all.
Not a word on her blog to the contrary. Not even a single letter of a word even.
If she didn’t read them, then OMFG WOW! Now that would truly be newsworthy RE:
“CURRY CAN”T EVEN READ!”
She claimed, in her first post on BEST results, that she had input into the writing of the papers.
OK.
Found it;
“My contribution to these papers has been in the writing stage and suggesting analyses.”
So then, having read said papers, and having contributed to the writing of said papers, Dr. Curry NOW takes objection to her own readings, reviews, and recommendations of said papers.
“I have not had “hands on” the data, …”
“The papers are initial steps in analyzing the data set, and the verdict on these particular papers will be given by others who do subsequent analyses.”
So, not having done any analyses whatsoever, and having passed on to “others” the “verdict” of “subsequent analyses”, she is of the mind to “move the goalposts” such that she can claim no discernible temperature trend, since when? Today.
Foot in mouth disease indeed. :-)
She also said that her 2nd author slot is due to alphabetical listing of the non-principle authors.
As a way of diminishing the importance of her role in the generation of the papers now that she’s decided to the project’s “tarnished”.
Uh, what does “the writing stage” mean? I can imagine many different definitions for this phrase; is there a customary usage (say, “everything up to the final editing and proofreading?”)
Don’t mean to be picky, but I really don’t know what she means.
I take “only involved at the writing stage’ to mean she wasn’t involved in data collection and number crunching. She was only involved once actual writing of the manuscript began, i.e., advice on what to include, how to word stuff, possibly contribution of a sentence or paragraph if it dealt with her field of expertise.
And the alphabetical ordering of second through final authors is no big deal. Not standard, but far from unknown.
Really, some of you guys are over-parsing this stuff.
Curry elucidates:
“curryja | October 31, 2011 at 10:16 am | Reply
I have stated previously that to my knowledge there are no honorary authors on the best papers. I saw and participated in extensive online discussion about the analyses and the papers”
Thanks for that suggestion, Steve S.; it sounds plausible, at least.
Well then, it appears that Dr. Curry read these four papers and had “extensive online discussion about the analyses and the papers”.
So then, it would appear that at those times in the past, Dr. Curry had no strong objections to what was included in said same four papers.
If Dr. Curry did had strong objections to what was included in the four papers, in past discussions, one would expect to see some proof of those past strong objections.
As Clara Peller would say, “Where’s the beef?”
Well then, we had “Climategate”, so I’d suggest that this one should be dubbed “Currygate”. ;-)
Tamino,
is there any way you can run a true “land-only” temperature composite using your method to compare to the BEST method. It seems that everyone keeps saying that BEST only correlates well with NOAA but ignore that GISS met only is not a land-only method making them not directly comparable.
Robert – I think Judy Judy Judy is now saying BEST has done an analysis allowing an apples to apples comparison, and that BEST preliminary is closest to GisTemp; as in, GISSIN’ cousins on land and yet to be determined at sea?
Clear Climate Code can be used to generate a GISS-algorithm temperature series covering land only: See http://clearclimatecode.org/a-real-land-mask/
Well if you think about her double negative:
“There is no scientific basis for saying the warming hasn’t stopped”
I think to paraphrase this more precisely you would say:
“You can’t prove it has warmed.” Or “The trend is not statistically significantly different from zero.”
But the way she has phrased it is misleading (perhaps deliberately) and makes it sound like it has not warmed. But the fact that we have to debate what she’s saying shows that at the very least she does have a case of foot in mouth disease.
‘There is no scientific basis for saying that warming hasn’t stopped,’
Is what I’d call a true but totally pointless statement. Global warming may have stopped 10 years ago, or a day ago, or a second ago. We can’t scientifically prove that none of these ‘stops’ occurred.
Another true statement is ‘There is no scientific evidence that global warming has stopped’
Or maybe I’m just playing games with words and we all know what Judith really meant. its pretty obvious what many ‘skeptics’ would interpret this statement as ‘The statement that global warming has stopped is scientifically reasonable’.
[Response: There’s a difference between “We can’t prove it hasn’t stopped” and “There’s no scientific basis to say it hasn’t stopped. The former is true, the latter is not.]
Seque into Tamino’s inline comment above.
This is why I don’t understand no apparent notice of Muller’s story that BEST and GisTEMP are closest as of now, with CRU being an “outlier.” It appears to me that all of the claims of flatlining/ hiatus/ pause/ shift/ global cooling/etc. are based upon HadCRUT. Kyle Swanson as much as said this on RC when he did the article on the Tsonis-Swanson paper – the shift sort of faded away on GisTemp then, and I think it’s vanished now.
Isn’t that precisely what the CERES TOA energy balance measurements do?
I did a count of the data that BEST has for 2010. In March they have 14488 stations reporting. In April and May there are 47. And in June, none.
And April and May are all Antarctic.
Correct me if I’m wrong, but aren’t those two months a bit cold Down South? :)
That would be irrelevant, as they are anomalies for those stations. Apparently, 2010 was a cold autumn on Antarctica. Heh. It was a warm spring on the NH.
Nick,
Could you provide some guidance for getting at that data? I’m sure this is going to come up some place that I wander.
The data are available at the BEST site http://berkelyearth.edu
Spelled that wrong, http://berkeleyearth.edu
And the internet reports faithfully “that’s a .org to you, buddy”!
Dave123,
I’ve put a copy here too. It’s in a much more compact GHCN v2 format. Even so, the site required me to split it into 2 11 Mb parts, bestdata1.zip and bestdata2.zip.
There is also a file ALL4inv.zip which has the associated inventory with names and metadata. It has .CSV files for GHCN, GSOD and CRUTEM3 as well.
Nick, I have compared yours to the original and it appears to be the same.
So I have to ask you the question, have you actually looked at the data values.
It has a lot of major problems, I was interested in looking at the data for Greenwich UK (station 148987) close to where I was born for around January 1947 when I was born.
The winter months have higher average values in 1948 than for most of the summer of 1948. This is a very common problem with dataset.
The other problem is odd minus values appearing within surrounding plus values, dropping averages by 30-40 degrees.
See sites 149254 & 149255 for examples.
Well, that pretty much settles it then. Those months clearly should not have been analyzed because of inadequate coverage of the data. Good catch.
Well, they did make clear the error bars were huge … not that the children have noticed, of course!
I’ve been coming back to these comments and the above graphs on and off all day… jaw dropping.
However. This is, IMHO, an excellent example of what peer review is actually really all about.
I read WTFUWTs breathless post, and it boiled down to two things:
First, instead of saying it hasn’t warmed since 1995, we should now say that it hasn’t warmed since 2001.
Second, Los Angeles isn’t warming like the rest of the world.
I must admit to reading the comments before I read the post, and I was seriously worried that the post might actually have something useful to say. But there was no need to worry……
Interesting that you state due to statistical uncertainty the rise per decade could be as much as 0.29C but don’t state it could also be a fall of 0.23C. Why not give the entire range of possible values rather than just one side of the range?
… as I said much earlier. ;)
Because, the underline long term trend since at least the 70’s is of an increasing persuasion.
Alas, gone emerita, as the saying is these days…
Well, I gave up on Curry’s blog after reading her “mind boggling” take on the Greenwire article so beloved of the denialati recently.
But a question that those more statistically competent than myself could help with.
There’s been a lot of blogosphere noise recently about the time necessary to show a significant trend in warming, Tenberth (I think) claiming 17 years is the time necessary.
I’d like to check my understanding of this which is:
If the underlying trend is as predicted, then 95% of the time the trend over the last 17 years will be positive.
Ergo, if we look at the 17 year trend for each of the last 20 years, we would *expect* to see an obvserved negative trend over 17 years once.
And over a century we’d *expect* to see an observed negative 17 year trend five times.
Is this broadly correct?
And if so wouldn’t it be better rather than to say that a 17 year trend shows significance to say that we expect a 17 year trend to be negative for 5% of the years in a given sample, or something similar?
Which, if true shows just how ridiculous claiming a “hiatus” in underlying trend is from a *ten* year trend.
Yes, about right. The precise period may be 14-17 years depending on which global data set you look at.
Dhogaza (1.29 am) It’s principal authors not principle authors. Just thought you’d like to know.
Or perhaps he meant to write:
“She also said that her 2nd author slot is due to alphabetical listing of the non-principled authors.”
Yeah, well, just wait until you make a typo … :)
I know it’s principal, not principle (complete with comma).
Between Curry and Rose I’ll believe whichever one can damage the other the most. Right now I don’t know who the unprofessional liar is. Convince me and the rest of the public!
MD – your level of cynicism is completely called for.
I believe Rose has posted again on Climate Etc., so he’s not backin’ down. Hell hath no fury like a journalist scorned.
When Anthony Watts will be (finally) ignored, climate will be (finally) destroyed…
Alex
And the beat goes on and on and on and on…. most important…..what will the earth be like for future generations if we choose to ignore the data we don’t like and only like the data that supports the stand we choose…
John, it turns out that warming is almost always in the process of stopping:
http://doc-snow.hubpages.com/hub/When-Did-Global-Warming-Stop
It seems to me that Judith’s comments reveal not just a lack of understanding of the data, but also a lack of understanding of how the climate works. Anyone who has studied climate for even a short while should know that there are natural fluctuations in global temperature which have little or no bearing on long-term trends – mainly the solar cycle and ocean oscillations. To have periods of a decade or even longer when there is no significant net increase in global temperature, despite a continuing long-term warming trend, is not in the least surprising – in fact it’s expected. Judith must either be *extremely* ignorant of climate science or she knows this perfectly well and is just being obtuse. There is no evidence *whatsoever* of even a slow-down in the global warming trend, in fact it’s still accelerating.
I wrote this on another forum in response to someone claiming ‘global cooling’:
Let’s do a proper analysis to figure out whether there’s any ‘cooling’ to be concerned about. We’ll use a period of cooling which everyone is familiar with, and no-one disputes, in the mid-20th Century –
http://woodfortrees.org/graph/gistemp/mean:12/plot/gistemp/mean:132
The warming period leading up to this cooling was from around 1915 to 1945 –
http://woodfortrees.org/graph/gistemp/from:1915/to:1945/mean:12/plot/gistemp/from:1915/to:1945/mean:12/trend
This gives a warming trend of 0.14°C per decade. We can see how this changes as the warming ceases and cooling kicks in. Look at the trend at intervals of 5 years:
http://woodfortrees.org/graph/gistemp/from:1915/to:1970/mean:12/plot/gistemp/from:1915/to:1945/mean:12/trend/plot/gistemp/from:1920/to:1950/mean:12/trend/plot/gistemp/from:1925/to:1955/mean:12/trend/plot/gistemp/from:1930/to:1960/mean:12/trend/plot/gistemp/from:1935/to:1965/mean:12/trend
1920 – 1950: 0.10°C per decade
1925 – 1955: 0.05°C per decade
1930 – 1960: 0.01°C per decade
1935 – 1965: -0.01°C per decade
So do we see any sign *at all* of this happening now? –
http://woodfortrees.org/graph/gistemp/from:1960/to/mean:12/plot/gistemp/from:1960/to:1990/mean:12/trend/plot/gistemp/from:1965/to:1995/mean:12/trend/plot/gistemp/from:1970/to:2000/mean:12/trend/plot/gistemp/from:1975/to:2005/mean:12/trend/plot/gistemp/last:360/mean:12/trend
No. The warming trend shows no sign of diminishing at all –
1960 – 1990: 0.08°C per decade
1965 – 1995: 0.11°C per decade
1970 – 2000: 0.16°C per decade
1975 – 2005: 0.18°C per decade
30 years to date: 0.18°C per decade
If there was any truth to the claims that global warming stopped in the last 10 years, or since 1998, or since 1995, or whatever the latest claim is, then we can see from the earlier 20th Century example that we would already be seeing clear signs of that in declining trends over the last several 30-year periods. In reality the trends are still increasing, not declining – i.e. warming is accelerating.
We can say for sure that *right now* the warming trend continues unabated, at around 0.18°C per decade, with no sign whatsoever of any decline.
> lack of understanding
Charitable today, are we
I did a sanity check testing all the 10 year trends in the satellite record.
What you notice is that there are no recent cooling trends. There were, though, a couple of 10 year trends in the early part of the data. My sense is we are about to embark on another rapid rise in the warming trend over the coming decade.
Buckle your seat belts boy and girls.
Rob, I think you’re right –
http://www.woodfortrees.org/graph/gistemp/from:1970/mean:48/plot/sidc-ssn/from:1970/mean:12/scale:0.002
Based on her lengthy experience she _should_ have known. Unless she is in denial.
I don’t know whether the short blurb I saw about Muller is accurate or what, but in it he’s quoted about his recent apostasy with words like, “Two years ago, a reasonable man had to be a sceptic.” So, Muller’s “conversion” is a bit of a preposterous slam at climate scientists everywhere. We’re supposed to be grateful that Muller has given AGW his Good Housekeeping Seal of Approval. Other people have taken a similar route long before he did but obviously they just didn’t do it correctly.
Hey, starting with Crichton was right and ending with GisTemp is BEST is a nice story.
i.e. he’s looking after an enormous ego.
Jeffrey Davis:
Yes, that was an odd quote and it’s in the wire piece that’s showing up in the press nationwide (including my local daily).
I attribute it to a combination of ignorance and his obvious belief that leading climate scientists have been engaged in nefarious behavior, because he’s speaking of “points raised by skeptics”. He obviously assumes that issues like the UHI were raised by skeptics (recently?) and have never been addressed by mainstream climate scientists. He obviously assumes that mainstream climate scientists have ignored siting issues (which is why, years and years ago, a new network of well-sited stations meant to provide better data for climate work was proposed and is being implemented, that’s the kind of behavior professionals ignoring issues exhibit, right?).
So, in other words, his history is that for years he believed every story he was told by the likes of Watts regarding the honestly and professionalism of climate scientists.
I get the impression he’s struggling with the results. On the one hand, he says that any reasonable person should’ve been skeptical two years ago due to valid points raised by skeptics (despite these points having been dealt with by mainstream climate science for *years*).
On the other hand, he has testified in Congress that gee-whiz, those climate science people have been doing an excellent job with their temp reconstructions after all, that they’re competent, etc. Though he did say they were surprised to learn that this is true …
All of which is why you’re supposed to read the literature *first*!
Hey, at least he did the work. Or he got Dr. Rohde to do the work. Or something/somebody. Anyway, it did get done. . . and apparently honestly.
Kevin McKinney …
Not physicists! :)
Or maybe Muller thought that WUWT is the premier online journal of the climate science community …
Can someone point me to a potted history of Muller’s involvement in climate science? I can’t believe that he really is so credulous and gullible (not at all sceptical in the true sense).
See Andrew Gelman’s take, including an interesting discussion from someone who attended a Muller talk. See Muller’s Global Warming Bombshell
A prime piece of evidence linking human activity to climate change turns out to be an artifact of poor mathematics., much of which was included in Muller’s book “Physics for Future Presidents” (2008), rather long after Wahl&Ammann and other information was well-known.
John I think this is the link you wanted to post:
http://andrewgelman.com/2011/10/skepticism-about-global-warming-skepticism-skepticism/
GP: thanks, yes, that’s what I meant. Sorry.
“Noise isn’t trend — by definition”
Yes OK, in one way. But not if you include natural variability in your ‘noise’ as you wrote shortly after.
But what is that supposed to mean here? Unless you are very very confident in the origin of each and all of the underlying contributing mechanisms, there is no way anybody can separate what part of the data constitutes a true trend and what supposedly is noise. None whatsoever! The data contains all causes (and measurement errors on top of that)
As Andrew P notes, the noise in actual recorded temp-data (especially when globaly aggregated) is much smaller. And the a fitted ‘trend’ is an a posteriori construct to characterize that dataset. It is a result of the data, it does not add any additional data or information. And neither the trend, nor its extension/extrapolation takes precedence over the actual observed data. Which are available.
[Response: You are very confused. Noise is random rather than deterministic, that it’s a random physical process rather than measurement error doesn’t change that.]
If I understand this post correctly, the main thrust is that the over the last decade observed fairly flat temperatures by a small probability still could hide a ‘true’ underlying trend that could be higher than 0.29°C (or just as likely lower than -0.23°C/decade)
But that is plain obvious for any set of random (or measured but fluctuating non-trending) datapoints.
[Response: You are very confused. The point is that the trend estimated from the last decade (or less) is not demonstrably different from what it has been over the last 30 years (or longer) — one which is certainly NOT zero (or below). If you want to claim global warming has ceased, you need to provide evidence that the trend is different from what it has been. Nobody has yet done so — not Judith Curry, and not you.]
Then you go on to construct ‘trends’ if you had started your interval earlier. And note that before that decade it was indeed cooler. There is no argument there.
But when doing so, starting ever earlier years, constructing new trends from then to now, you show that a fitted trend has significant positive slope if you start from ~1997 (or even the ~90ies for higher uncertainties)
But that your trend calculation is not significant (does not exclude zero, ie notrend) from there on!? And your whole point to start with was that the uncertainties could indeed hide a more positive trend (but also a more negative one)
I have an idea why some people are so eager to talk abot the ‘trends’ and elaborate around them, when the actual data is available, and everybody can look at it …
[Response: I have an idea why you’re so confused.]
Tamino, send this guy to the sin bin. He’s been awarded his own thread at Deltoid and should be confined to it.
Seconded.
I agree too; nuke him in the bud. His thread at Deltoid usually dominates the “new post” list, swamping out other discussions. He never learns and never admits a mistake; he’s a site killer.
[Response: Done.]
“I have an idea why some people are so eager to talk abot the ‘trends’ and elaborate around them, when the actual data is available”
Is there something I’m missing? Trends are a particular way to describe data.
“But that your trend calculation is not significant (does not exclude zero, ie notrend) from there on!? ”
No, this does not mean there is no trend. A statistic textbook will help you.
“A statistic textbook will help you.”
I wouldn’t count on it–unfortunately.
Ah, Jonas, making like a bad penny I see…
Tell us – how many ‘flat’ (and ‘declining’) 10-year periods are there in the temperature record for the last century of data? What does this mean?
And can you address the post here?
Deepclimate wrote, ‘What’s the point of a modern temperature record that is more than a year out of date?’
Being already out of date (and hard to update) is a drawback IF the point was to create an ongoing time series, an alternative to GISTEMP, HadCRU and NCDC.
But keeping up to date does not matter if the point was to show that GISTEMP, HadCRU and NCDC had got it wrong historically. From Muller’s pronouncements, that’s the goal he had in mind, and was surprised not to achieve.
[Tamino’s Response: I have an idea why you’re so confused.]
but it’s too large to fit in the margin,
—Pierre de F
outch :]
The globe has cooled since 1973. All trends since that time are negative. If all trends are negative, then the temperatures now must be less than they were in 1973. My scientific proof is here. How can anyone argue with this?
Looks like an open-and-shut case to me.
jeez…who could possibly argue with such an overwhelming offer of proof? six overlapping trends has to be some kind of New World Record, eh?
Now that’s what I call creative charting! :-)
Nice Poe.
Hahahaha, sweet! I may pinch this.
Yes, but the cooling trend is slowing. I notice that the slopes of your cooling fit lines steadily approach zero. :)
Instant classic!
This was foreseeable, was it not? Release Muller’s statements so blogosphere & media will use the “skeptic changes mind” news novelty to elevate EST’s volume; then let Muller or another redirect the now-prominent discourse to different ends. Like doing a flyby to have planetary gravity sling your spacecraft off in a different direction.
(Apologies if others have pointed this out; many comments, & I don’t know what keyword to search for.)
…or maybe not; Curry says “the reason for the publicity blitz seems to be to get the attention of the IPCC. To be considered in the AR5, papers need to be submitted by Nov, which explains the timing. The publicity is so that the IPCC can’t ignore BEST”
Rose has “form” as they say in the UK… Deltoid has a whole series on Rosegate. He seems to be devoted to Steve McIntyre.
http://scienceblogs.com/deltoid/global_warming/rosegate_1/
Nice to know that the Daily Mail still retain the services of this trustworthy journalist, and do not apply extra factchecks when he reports climate science.
Re Jonas
I think you are having a hard time grasping the strict application of the statistical test being used. The point is to test whether recent short-term trends are inconsistent with the long-term underlying trend given all of the natural random variation and measurement error that occurs around temperature trends. You are right that the results of this cannot be interpreted any more broadly than that specific question. But that was never the intent.
That’s the only way you could prove from a purely statistical standpoint whether recent trends were somehow inconsistent with long-term warming.
Alternatively, we could try and remove some of this random natural variation using known causative correlations. For example, we have good estimates for how ENSO and TSI affect temperature. Tamino and others (including myself) have done exactly that elsewhere. And the results show that after removing natural variation caused by ENSO and TSI, the underlying trend has remained unchanged over the last decade near .18C/decade.
OK, hands up anyone who is surprised by the behavior of either Curry or Rose. When was the last time either had anything intelligent to say. I’ve looked back into the dark and distant past with no luck.
Not surprised.
I just wish that they (along with a few other septics) would lay their email inboxes open to the public. They seem to consider such things fair game, and given their unimpeachable honesty and integrity I doubt there would be any embarrassing inclines to be hidden.
This is really sad but indicative of how cunning and successful the deniers have become. No evidence against them can be useable for very long.
When some does come up, concocted by one of them, they unleash a storm of noise and confusion apt at bewildering most reasonable people in the general public. It does not matter how idiotic or nonsensical Curry and Pielke can be on statistical significance, the very fact they’re out there talking is enough to add to the confusion.
In the perception of a public that is often barely numerate and ideologically pumped up on a daily basis by outlets like Fox News, the attempts by reality-based blogs to sort it out do nothing but add to the noise. All they take out of it is that there is “controversy.” Then they of course retreat to whatever opinion is safe for their ideology and world view.
This is a mind manipulation war of the highest degree. It is unfortunate that the public’s level of math literacy and overall critical thinking skills have created such an easy terrain for the manipulators. The more I witness this happening, the less I see what effective measures can be taken against it. It’s too easy for them. The moment they’re caught with their pants down they start screaming and squirming and that’s all it takes to distract the gooffy ADHD child known as public opinion.
This vidoe presentation was mentioned over at Skeptical Science;
Dr Richard Milne – Critical Thinking on Climate Change: separating skepticism from denial
[edit: please no embedded video in comments]
IMHO, quite well done as far as a “classroom” lecture goes.
Now, if only the “climate disinformation campaign” could be presented to the broad public in as clear and concise a manner.
Louder and more repeatedly than the .”climate disinformation campaign” is to begin with. One would hope.
Sorry about that. When I posted it I saw that it was embedded, I do know your site policy on embedded videos, my mistake. I too dislike embedded videos in discussion threads.
How about a link to the Skeptical Science article instead, that has the video?
http://www.skepticalscience.com/Richard-Milne-separates-skepticism-from-denial.html
Thanks
Phillipe,
You’re absolutely right. How much longer will we hear “See? There’s controversy … no consensus!” And Rick Perry will be saying that “they laughed at Judith Curry … just like they outvoted Galileo”
I know I’m always the optimist here, but I do think that the idiocy of is a little more widely self-evident than we sometimes think.
“There is no scientific basis for saying that warming hasn’t stopped.”
There are enough negatives in this statement that she could try to find wiggle room by saying she wasn’t actually saying anything at all.
One can’t demonstrate that warming hasn’t stopped, starting five minutes ago, for example. Though one also has no conceivable reason to claim that it has stopped warming as of five minutes ago, unless one specialises in spreading confusion.
‘There is no scientific basis for saying that warming hasn’t stopped’
If I said that this statement was about as true as:
“There is no scientific basis for saying that faeries do not exist”
Then I think we would have agreement?
I’m wrestling with whether I believe the statement about faeries is true or not, and at the moment I’m leaning towards considering it a true statement. In my mind its not science that says there is not such thing as fairies, but rather science says there is no evidence for fairies, and common sense says that if there is no evidence for it, it probably does not exist.
That’s about what I’m saying. Taken in isolation of her other comments, she didn’t say anything at all. I can’t prove that fairies don’t exist; I can’t prove that leprechauns don’t exist; I can’t prove that warming did not stop in the last x years, where x is shorter than the period of time needed to detect the long-term signal from amidst the short-term variability.
But I also don’t have any good reason to say that any of those things do exist. And it’s also a bit of a dog whistle; we’re parsing the statement carefully and literally, whereas other sorts of readers will happily take it as somebody with credentials saying that warming has stopped. And then you’ve got the other things she’s said, or at least has been reported as saying. Finally, given her track record of speaking to issues she hasn’t looked into in any depth, one might not be inclined to give her the benefit of the doubt.
Semantics 101? You decide.
1) There is no scientific basis for saying that warming has not stopped.
2) There is no scientific basis for saying that warming has stopped.
3) There is (a) scientific basis for saying that warming has stopped.
4) There is (a) scientific basis for saying that warming has not stopped.
But if we take out all those weasel words and trim it to it’s barest minimum;
1) no science | warming
2) no science | no warming
3) science | no warming
4) science | warming
So I would argue that (1) and (2) would mean that without science we are left with a subjective opinion.
So I would argue that (3) and (4) would mean that with science we are left with an objective opinion.
I did learn a new word today, litotes (from wiki);
In rhetoric, litotes is a figure of speech in which understatement is employed for rhetorical effect when an idea is expressed by a denial of its opposite, principally via double negatives. For example, rather than saying that something is attractive (or even very attractive), one might merely say it is “not unattractive.”
Litotes is a form of understatement, always deliberate and with the intention of emphasis. However, the interpretation of negation may depend on context, including cultural context. In speech, it may also depend on intonation and emphasis; for example, the phrase “not bad” can be said in such a way as to mean anything from “mediocre” to “excellent.”
Tamino,
I hate to put a load on you, but McIntryre is throwing doubt on the “scalpel” used in the averaging method paper. It feels like BS to me, but I don’t know. He seems to be claiming that the change point analysis is wrong based on someone else’s (Menne) paper using a different method. I’m math impaired, only got through 2nd semester calculus and have forgotten most of that since it was 30 years ago and I never used it in real life, but it just seems wrong to me. Wherever the cuts occur should not matter since the analysis would just make appropriate adjustments.
We love you, Tamino!
Tamino,
A response from Curry popped up very breifly.
Judith said that she had no idea what you were asking and couldn’t understand your “screed” of a post.
Comment has now been removed!
Tamino, you might be able to turn off auto-embed of videos so you don’t have to keep scolding people for doing something they didn’t actually do. ;-)
http://codex.wordpress.org/Embeds
The money shot is here: http://codex.wordpress.org/Settings_Media_SubPanel
[Response: I did exactly that. It doesn’t seem to stop wordpress from embedding them anyway.
And I don’t think saying “Please no embedded video in comments” constitutes “scolding.”]
“Scolding”- perhaps a poor choice of words.
I suspect that “we” all know Tamino’s video policy and have no problem with it. I also suspect that we would all rather have Tamino apply his time to the statistics he does so well and not waste his time on an idiosyncratic administrative bug at his site.
Maybe an HTML hyperlink might get around the bug (rather than simply posting a video related URL)?
Whatever. The bug is here. Tamino has already tried your tip pough (but thanks anyway). As usual he is already a step ahead of most of us.
Having preview would help
Yes GP,
Being somewhat HTML impaired, I have, on occasion, used other’s “preview” feature (thanks SkS). It helps, but not with this problem. It is also time consuming and leaves one open to the possibility of posting something really stupid looking somewhere else.
There is also an issue of merely seeing a longish post in the typing window here.
It’s an acid test I guess.
The professional deniers won’t let up until Muller gives in. Does anyone have news from Santa Fe?
http://www.cvent.com/events/third-santa-fe-conference-on-global-and-regional-climate-change/custom-19-bb57a62c4a1f40d8af285205963998df.aspx
Definitely the place to be these days…
A poster on SKS linked to this website which details climate scientists explanations for what they term the warming hiatus of the past decade.
http://www.eenews.net/Greenwire/2011/10/25/1
Tamino, Jonathon is none other than Eric the Red / Dana Hicks / Hans, among his many internet identities (all 4 have been censored for numerous violations of the Comments Policy at Skeptical Science).
The poster that Jonathon thus links to is none other than himself (see here and here) in one of his guises. Part of the game he plays.
It is recommended to consign his comments to your equivalent of the Bore Hole, for he will continue to prosecute his agenda of disinformation and dissembling here, degenerating the dialogue into a morass of corrections and repetitive postings.
Feel free to delete this message and/or email me on this.
[Response: Anonymity is fine. Sock puppetry is not.]
Maybe a bit OT but this 14 or 17 year trend has had me pondering if this has any bearing on the behaviour of Periodic Cicada where the 17 year periodics are generally found in the northern hemisphere and the 13 year periodic cicadas in the southern – with some overlap. Both types can be found in the same forest. Any comments anyone?
Never mind the sun, its the cicadas wot done it. ;-)
Whatever a most interesting topic so keep up the good.
Nah, periodicity is a very different thing from the time scale on whih something becomes statistically significant… I couldn’t imagine a mechanism for any such relationship.
Judith Curry protests that she was misrepresented by the article in the Daily Mail/Mail on Sunday. What Judith has actually said on her blog is:
“To set the record straight, some of the other sentiments attributed to me are not quite right”.
Not quite right eh? But basically OK? So, why did Judith write later:
“Well, I have a rule about not talking to reporters on the phone, asking for submitted questions and I respond by email. It’s a rule I extremely rarely break, and Rose caught me on the phone and I spoke with him.”
Judith is obviously savvy enough with regards to the ways of the press, to know she should have this rule, and, yet, when she did choose to break it, she tells us that it wasn’t with the BBC or Washington Post, with whom we could all believe she inadvertently might, but actually with that most scumbag of all outfits the Daily Mail.
Does this all ring true? Call me cynical if you like, but if Judith had wanted to create a measure of uncertainty and doubt in the popular mind about the BEST findings, then allowing the Daily Mail to do their worst with her comments would be a neat and deniable way of doing it, don’t you think?
Of course the problem is that interpreting all of this is so subjective for the public.
I’m having trouble understanding how two scientists involved in the same study seem to be characterizing the results differently, the results of which are not in dispute (yet). This is ridiculous, couldn’t they talk to each other before undermining their own work publicly? I’m no scientist, but making strong statements about climate change conclusions in as short a period as 10 years seems so anomalous. What’s wrong with both of these scientists for not being more clear and offering a consensus to the public especially given the raucous environment? All I can think is Muller was blindsided by Curry and this has undermined both of them.
Andy
This actually was an email response to a friend.
On Rose’s blog ““…But I think your memory is at fault when you state that it was I who first used the phrase “hide the decline”. You did this, twice, in our first conversation, although it’s true I was the first to mention it in our second talk…”
Whilst Curry says “Well, I have a rule about not talking to reporters on the phone, asking for submitted questions and I respond by email. It’s a rule I extremely rarely break, and Rose caught me on the phone and I spoke with him.”
So she ‘extremely rarely’ breaks her rule but did so at least twice with Rose of all people, and then implies they only had one conversation. Hmmm…..
The woman just won’t stop. Now she’s hawking two papers, one published in E&E one published in International Journal of Modern Physics C (a sister journal of the one which published G&T) by authors associated with CFACT, the group which pays Mark Morano’s salary.
Curry justifies her choice of giving attention to two papers from people from a clearly politically motivated group (EIKE):
“curryja | November 7, 2011 at 8:45 am | Reply
An argument is an argument. Given that Ludecke’s arguments got published in reputable scientific journals, their merit (or lack thereof) should be discussed.
I am not worried whether or not you or anyone else can figure me out. But its really very simple. I don’t play political games with the science. Dismissing someone and their published work because they belong to a certain political group is playing political games.
There aren’t too many climate blogs that don’t play political games with the science, which is why my not playing political games with the science seems like such a foreign concept in the climate blogosphere.
I’ve abandoned trying to build bridges with the climate dittoheads in the blogosphere. I’m trying to engage the thinking public to actually think, and challenge their understanding and prejudices.”
That was a pretty remarkable comment by JA. Not sure I would describe her acolytes as the “thinking public.” I certainly don’t see her challenging their prejudices.
Curry famously got flack at RC for recommending Montford’s book without having read it, but now she’s praising her own blog without apparently having read it – that really is something.
IMO it can only be coercion.
Could we have another Congressional hearing please, and ask her under oath?
This is some excellent work you’ve done here, and I not only applaud your rigorous and thoughtful analysis, I make the same argument here, using some of your analysis.
Whoever you are, tamino, keep it up. I’ll keep standing up for solid science and against disinformation and fraud wherever it happens.
Ethan, I note that some of the dimmer bulbs who comment on your blog bring out the old “what is Earth’s perfect temperature” meme.
My response is that it is that temperature which allows us to feed, house and clothe 10 billion people by mid century. I then point out that agricultural yields mostly go down with temperature.
You did a nice job.
You don’t need to dig too deep to see this whole thing is political, and shameful that a university as good as Berkeley (and indeed scientists with the track record of some involved) should stoop so low. These people are trying to ride the “climategate” story to turn themselves into the heroes who once and for all made the fossil fuel industry’s case, except their numbers don’t support it and some of them are (just) honest enough to say that. The real dishonesty is in characterising previous scientists’ work as inadequate when they have in effect shown the opposite. They’ve even demonstrated that, though it would have been preferable had CRU made every last detail of their data public, it doesn’t matter. It’s not like we are analysing trade secrets or something of that kind. Data about the planet will be mirrored in other data sets and other measurements, and what they have demonstrated is that you can start from scratch and get the same answer. Nothing to see here. At least 3 other climate science groups already do that, and regularly publish their results, some with all data and code freely available (e.g. NASA).