Friday, December 11, 2009

To scare or not to scare, that is the question

The recently released Copenhagen Diagnosis assessment has been accomplished by 26 scientist, down from 4000 or so that contributed to the Fourth IPCC Report. These 26 have been described to be 'leading scientists', raising the question ‘what are they leading us to’?.

The assessment is theoretically based on peer-reviewed literature, and thus was not intended to include new research. It aims to just up-date the findings of Working Group I (The physical basis) of the IPCC summarized in the Fourth IPCC Report. The Copenhagen Diagnosis may be indeed be based on peer-reviewed studies, but it soon becomes clear that it is not based on all peer-reviewed literature, and not even on the most relevant peer-reviewed literature. The selection of published papers that the authors had considered worthy of assessment is indeed quite worrying.
Let us consider the projections for sea-level rise for year 2100. The Copenhagen Diagnosis indicates that global sea-level could rise up to 'at least twice as large as that presented by IPCC AR4 with an upper limit of 2 m' . There are, indeed, published studies that estimate a sea-level rise of that magnitude. For example the one by Rahmstorf (Science 07), solidly founded on a statistically robust regression analysis based on 4 to 6 degrees of freedom. Rahmstorf's results are quite prominently displayed in the Diagnosis chapter devoted to global sea-level rise. It is, however, disturbing that results of just one team are prominently display in a figure in a climate assessment written by that very same team. We are now suffering the consequences of this hockey-stick-strategy implemented several years ago. As a rehearsal of a deja-vu, two recent papers about global sea-level projections published in Nature Geosciences are simply not discussed and not even cited. Why?

The first reason we must consider, of course, is that the excluded research is of poor quality. The inclusion of the work in Nature Geosciences presumably makes that reason highly unlikely (although I do note the rather high percentage of papers published in Nature and Science  seem to later be disproven). It is also of note that the authors of the excluded papers have long publication records on global sea-level, and one is the coordinating author of the next IPCC report.
Having ruled out the quality of both papers and authors as being overly problematic there is no option but to turn to the actual contents of the two papers. Both abstracts state clearly that global sea-level rise in year 2100 will likely remain under 1m, and thus contradict Rahmstorf 07. Two critical comments to Rahmstorf 07, along with Rahmstorf's responses, were also published in Science in 2007. These critical comments have been ignored in the Copenhagen Assessment as well.
Polls indicate that public conviction about the existence of anthropogenic global warming is wobbling. I think that this strategy of 'hide-and-scare' was and is becoming the most serious threat for the IPCC and for a rationally agreed solution. It is the time for surgery that we are running out of.


Stan said...

Given that Rahmstorf has been clearly established to have suffered from gross incompetence (if not outright fraud), why would anyone give credence to anything he might say?

Anonymous said...

About half of the Copenhagen Diagnosis authors are IPCC authors. The IPCC is only required to deal with human-caused changes, so the IPCC authors are used to ignoring natural variations.

Are any of the Copenhagen Diagnosis authors members of the Climategate web?

bernie said...

Your approach is very helpful. Informed and temerate voices are very welcome.
I sometimes think that people simply cannot deal with large numbers or long time periods. 100 years ago the Zuyder Zee was an unpleasant and dangerous place. Man has a tremendous ability to adapt to actual events. The modest probable increases in sea level over the next 100 years is definitely a threat to large numbers of those living in and around the major SE Asian deltas. But it is manageable and, with modest international help, addressing this reality via dykes, pumps and the like may lead to the development of more functional societies.

Stan said...

See also

"So, the short criticism of Rahmstorf’s method: Stefan used a bad method (smoothing) and then applied it in the region where it is especially bad.

As a method to test projections, the method Rahmstorf selected remains a foolish choice."

Bottom line -- he averaged made up numbers with the real data to raise the smoothed graph and then declared it was "worse than we thought".

Amazing how those pretend numbers were worse than the real numbers. And it proves we're all going to die!

richardtol said...

I'm with Stan on this one. When I discussed Rahmstorf's paper on sea level rise with him, he was not aware of vertical movement of the land, let alone its impact on measuring the level of the sea.

richardtol said...

To answer the question in the title, scare tactics work to mobilize the population against an outsider for a short period.

Climate policy is for the long haul, and all of us emit greenhouse gases.

Vinny Burgoo said...

A Richard Tol: 'When I discussed Rahmstorf's paper on sea level rise with him, he was not aware of vertical movement of the land, let alone its impact on measuring the level of the sea.'

When and where did this discussion take place?

(Have you thought of using another moniker to make it clear that you are not *the* Richard Tol?)

Nanonymous said...

"solidly founded on a statistically robust regression analysis based on 4 to 6 degrees of freedom."

LOL. Good one!

Anonymous said...


a good analysis.
Would it be time to also discuss the rather important role which Nature and Science have played in particular as highly visible and therefore influential "science tabloids"?

First, as you correctly point out, a significant number of papers have appeared in recent years in these journals which were technically flawed, or highly speculative (good maybe for discussion in a specialized journal which allows ample discussion of caveats, but not good in the format imposed by Nature or Science), and - since they didn't hold up to scrutiny even at the time of publication - should not have passed the review process (who am I to claim this, so arguably a matter of debate).

Second, I think it's fair to say that a robust but unexciting paper which demonstrates absence of significance of a signal or absence of trends from available observations is not the sort of thing of interest to these specific journals. Rather, submissions of papers which will likely make a "splash" (impact factor, right?) are highly encouraged (again, a claim which the editors of these journals will likely disagree with). The temptation to see a signal where there is none (or none of significance) is all too big.

Maybe the community should come to a consensus to drastically downgrade the value of these journals to encourage colleagues to publish in specialized journals aimed at thorough discussion among peers (note that I am not arguing against speculative papers, but it seems more usefully targeted at an audience which is trained at assessing the context and the caveats).

bernie said...

Good post but it would be helpful if you chose a more distinctive username.

richardtol said...

My discussion with Rahmstorf was a some workshop, I believe in Berlin. Never documented, so you'll have to take my word for it.

As far as I am concerned, I am *the* Richard Tol. I believe there are four other people by the name, and they would all argue that they're the real Richard Tol. I'm the only one who works on climate. I don't play judo.

Nanonymous said...

Re: Science and Nature as "tabloids"

Good observation! Here is the crazy thing - and I don't think there EVER was any scientific field with such a record.

The term "Global warming" was coined in 1975 in a Science paper. Since then, up to and including 1990, ISI Web of Knowledge finds 188 papers, reviews and articles containing "global warming" in the title or abstract. And get this: of these 188, the record is split this way:

By source
New Scientist - 38
Nature - 33
Science - 16
Chemical Engineering News - 13
Every other source - 5 or less.
(I.e., these four make up 53% of total!)

By publication type
Editorial material 91
Article 52
Letter 27
(I.e., the actual scientific publications make up ~28% of total!)

Seems like the global warming science was uncovered by journalists, isn't it?

For a comparison, in a SINGLE YEAR following this period, 1991, beat the entire decade preceeding it. There were 266 papers, reviews and articles containing "global warming" and 57% of them is listed as "article".

For further comparison, in 2008 alone, there were 1413 papers published and 77% of them were articles.
Reading some of those early papers, one can easily find clear evidence that from day one, the idea of

anthropogenic global warming due to GHG was taken up essentially as an axiom and that neither the
hypothesis, nor research conclusions have changed appreciably in the last 30+ years of "mainstream science".

Vinny Burgoo said...

@Richard Tol
Sorry, that was weird and presumptuous of me. It just seemed such a strange claim. It still does in the, er, sober light of a new day. A celebrated expert on oceans not knowing about how vertical land movements affect sea-level measurements? I need a drink.

richardtol said...


Rahmstorf trained as an oceanographer. He knows about currents and stuff. The coast is of no concern to ocean people.

Anonymous said...

Honest Broker... hm

I read some posts in this blog, until now, I cannot find the honest broker. Where are they, the honest brokers? Sorry.

Reiner Grundmann: references to Watts, Morano, dirty fraud accusations, criticless, dumb linking to articles of other blogs, typical "skeptic" tactics. Worthless.

von Storch: advertisements for his interviews. Boring. I know them.

you: Rahmstorf bashing. Boring, couldn't care less.

I was interested in this blog and I am deeply disappointed. You are not an honest broker.

Try, for example, this:
* review current research results from your point of view, put into contexts
* write critics to recent media reports
* show biases and politics with help of facts

ooops, realclimate does it, klimalounge does it, primaklima does it. I would be really excited to see your point of view because you claim you are the honest broker. BUT: vendettas, dumb bashing like Richard Tols and stupid links to Anthony Watts are boring as hell.

Show your opinions, show it with facts. Refute arguments made by stupid "skeptics" like Eschenbach, reply to stupid remarks that it is cooling, show that 2m is too high. Show the problems and the good sides of papers. Explain your interviews instead of repeating them, put them into context. Show the bias of some journalists, inclusively Ulli Kulke. But that would be work, that would need courage.

Until now, this blog is not good.

Of course, this is only my point of view as a disappointed layman. Maybe, I am biased and do not understand your point. Possible.

theoldhogger said...

To Anonymous
You are most assuredly biased. I recommend that you return to Realclimate and Joe Romm, etc......

Hans von Storch said...

Eduardo, why did you not mention our own analysis which showed that the Rahmstorff Ansatz does not even work in the simplified world of an AOGCM?

von Storch, H., E. Zorita and J.F. González-Rouco, 2008: Relationship between global mean sea-level and global mean temperature and heat flux in a climate simulation of the past millennium, Ocean Dyn. doi 10.1007/s10236-008-0142-9, 10pp.

Hans von Storch said...

To anonymous- maybe I call you Walter.

Walter: Thanks for your frank critique. This is certainly helpful, as we have to learn how to deal with our weblog, which is not even 2 weeks old. In general we will not attempt to correct our commentators (you), and our opinions you have to infer from our individual postings. After all, we are separate individuals. We will also not try to claim that we would know the final answer to contested questions.

Now, when you say I would just post boring interviews of myself; true – but I added a warning, didn't I? You do not need to read all postings; but what about the two quotes of Anders Fogh and of Pachauri? Was that interesting?
When I use the term "honest broker" this has very much to do with opening the debate; in inviting different opinions and views, without me or anybody else saying which of these is best or even true. How do you understand the term "honest broker", Walter? What do you expect? How would you write about things? -- Hans

@ReinerGrundmann said...

Walter -not sure how you would like to be called -

You cite a few weblogs you like and state that you do not like ours. Fair enough. If you think Realclimate is not biased, that's your opinion. Part of the blogging fun is to spread information quickly so people are aware of other debates and arguments going on.
Here is an example of a public event at MIT with Kerry emanuel and Richard Lindzen, among others:

And here is one that summarizes quite nicely how the fixation on 'the science' has been very unproductive:,0,3859887.story

EliRabett said...

Let Eli put it a different way, Ben Santer made four points in his statement on the current situation:

* We know that human activities have changed the levels of greenhouse gases in the atmosphere.

* We know that these changes in the composition of the atmosphere have had profound effects on Earth’s climate.

* We know that the human “fingerprint” on climate will become ever more visible over the next few decades, and will impact many aspects of our lives.

* We know that we are at a crossroads in human history. The decisions our political leaders reach in Copenhagen – or fail to reach – will shape the world inherited by future generations.

Which do you disagree with?

(Richard Tol might chime in on this also).

eduardo said...

I would disagree with the fourth assertion. The Rio summit decided, almost 20 years ago, on the limitation of GHG emissions, trying to set up quotas for each geopolitical block. These limits were later justified by some reasonable scientific assements, but also by some unjustified assertions (e.g. hurricanes, Katrina and the like).

Emissions have continued unabated, more or less the same as if Kyoto had not existed. Isn't it reasonable to question if the whole strategy is correct? Can the planet be saved by more Copenhagens, and only by more Copenhagens? what is going to change from now until Mexico?
I argue that the proportion of population that is convinced of the immediate threat of climate change has reached a peak.
Of course you can blame an oil-industry conspiracy that has hijacked people's minds. Intelligent politicians would, however, ask themselves how they could do more than preaching to the choir

Anonymous said...

The first paper is flawed, as aknowledged by the authors themselves. The correction brings the numbers close to Rahmstorf's:

The second paper, that doesn't present any original research, just don't say what you are implying, Zorita. They are speaking about the most likely value, not about an upper bound:

"A 'headline' figure of 1 m during the twenty-first century represents only the global average sea-level rise [...] Since then [IPCC AR4] several studies have suggested that a rise larger than 1 m cannot be ruled out."

Just exactly the same as what the Copenhaguen Diagnosys says: "sea-level rise until 2100 is likely to be at least twice as large as that presented by the IPCC AR4, with an upper limit of 2m". As anyone following the science perfectly knows. It is a very disappointing that you have decided to risk your reputation by joining the game of disinformation.

eduardo said...

@ 22
1. I would like to see a public correction of the Siddal et al paper. A comment by Eric Stieg in realclimate starting with ' as far as I am aware...' is really not enough. Realclimate has unfortunately not the best record of accuracy and unbiasedness. Two comments puting into question Rahmstorf's paper were published in Science (Rahmstorf responded to these). They are not cited either.

2.The paragraph in the Milne et al paper you are quoting ('a headline figure of 1 m ..' )explains that this is the global average value, and that regional deviations may occur. It does not indicate that 1 meter is the average global value of all studies. Actually, their abstract states clearly ' most studies constrain global average sea level rise to less than one meter over the 21st century'. The Milne et al paper is a review published in 2009. My question is: why is this review paper or any of those studies not cited in the 'Copenhagen diagnosis', and instead of this only one time series is displayed in the sea-level chapter?
I am not saying here that the Rahmstorf study is wrong. I am however indicating that the CD simply ignores all other studies that contradict Ramhstorf, and therefore is not a fair review of the state-of-the art.

3. Of those 'post-IPCC studies that do not rule out more than 1 meter of sea-level rise', one is Rahmstorf's itsef. So we have just one independent study (Pfeffer et al.) that may agree with Rahmstorf. All others do not. I guess that if you were a politician taking decissions you would like to know this, dont you think?

Anonymous said...


1. A lot of wrong papers are not corrected, they are simply abandoned. My point is that there's no reason to make things up: a co-author of the CD explicitly explained that the reason why the paper is not quoted is that they consider that the paper is wrong, and they informally explained why.

2. Given that the abstract is just a summary, please, let us know where in the paper they mention 1 m. as a constraint to the upper bound and what spcific studies they are referring to (most post-IPCC studies that constrain the upper bound of SLR to less than 1 m.). I cannot see support for that interpretation anywhere in the paper, I just see that they explicitly say that 1 m. is the global average, i.e., the most likely value (even Rahmstorf give less than 1 m. as the most likely value). The abstract was certainly clearer before the last-minute change, as they explicitly mentioned "most likely": "the majority of studies suggest that global mean sea-level rise will most likely be less than 1 m over the 21st century" (

There are now four papers supporting the summary of the CD, and Rahmstorf is the father of this semi-empirical approach:
1. Rahmstorf
2. Pfeffer, Harper & O'Neel
3. Grinsted, Moore & Jerjeva
4. Vermeer & Rahmstorf

It does seem that other researches think that the method is robust.

You also say: "The Copenhagen Diagnosis indicates that global sea-level could rise up to 'at least twice as large as presented by IPCC AR4 with an upper limit of 2 m'. There are, indeed, published studies that estimate a sea-level rise of that magnitude. For example the one by Rahmstorf (Science 07)".

A bit misleading, again, given that the upper limit of 2 m. is not supported by Rahmstorf 07 (he gives 1.4 m as the upper limit) and that doubling the numbers of the IPCC is supported by the various papers I've mentioned before and I'm not aware of any post-IPCC paper giving numbers as conservative as the IPCC's.

Hans von Storch said...

24 - it is interesting to see that some responses are rather late; they are welcome nevertheless. Of course it would be nice if they would be signed by a name, when requests such as "let us know" are included. But that is a matter of style.

But, Mrs 24, please have a look at the two comments, which Science published on this matter; the response was rather weak. One of the two comments, the one from Schmith et al. (Comment on ‘a semiempirical approach to projecting future sea-level rise’. doi:10.1126/science1143286) demonstrated that the statistics were not ok. Methodically not ok. In a nutshell, it was an effort to compare a trend with another trend. This is a field, where the co-integraqtion technique, developped in econometrics, is applicable, and the comment came just from such quarters.

One could then argue that one would have superior physical insight that the link would be a reasonable approximation; then it should be an approach, which works fine in millennial simulations. we have shown that it does not, not even in a simple model-world, where there is only thermal expanion. (this paper is available as open access: von Storch, H., E. Zorita and J.F. González-Rouco, 2008: Relationship between global mean sea-level and global mean temperature and heat flux in a climate simulation of the past millennium, Ocean Dyn. doi 10.1007/s10236-008-0142-9, 10pp.)

This paper was originally submitted as a comment to Science, but science rejected it because there had already been two comments. A strange argument to begin with.

Anonymous said...

25 Hans Von Storch,

My name will not tell you anything about me (Jesús Rosino, to satisfy your curiosity) and my anonymity shouldn't be an obstacle to address pertinent questions. I am not asking for a favor with requests such as "let us know where", it is just a rethorical way to assert that there isn't support for that anywhere in the document. I am late because it is now that someone has pointed out to me that the CD is crap because Zorita has purportedly shown that it is maliciously hiding results that contradict Rahmstorf's on projected SLR. And that was the specific subject to which my comments were restrained. Now you want to move the subject to assessing the merits of the critics and Rahmstorf's response. Well, I'm not really qualified to do that, but, as I said before, it seems to me that other researchers are going along with that method and developing it. The science is moving in that direction, so I guess that the critics were not that covincing. Regarding your paper, it seems that the contribution from Vermeer has achieved to improve that, hasn't it? (where they also say that R07 by design was not appropriate for such a comparison).

Hans von Storch said...

Thanks, Jesus Rosino.

I read from your mail that we would agree that there may have been a problem with R07. But that many people go along with it. Maybe because the critique is not taken seriously by R, and not mentioned in CD? We had such a case before, the hockeystick claim, which is now widely accepted as an early interesting attempt - which needed considerably revision. Another case was the case of the breakdown of the giulf-stream, which is still wildly popular in the public.

What do we learn from this:
a) The allegedly "top" journals "science" and "nature" have every now and then a serious problem with their review process - just like Climate Research back in 2003.
b) The idea behind R07 was premature; submitting a manuscript about premature ideas happens all the time; no fault on the side of the authors, but that is what the peer review process should take care of.
c) The case demonstrates the functioning of science - of self-repair, which needs time. The newest results are certainly not always the best, but those, which have been contested least.

The question is what conclusions lay-people should draw from this case. And what honest scientists? That is were Eduardo Zorita's comment here on the Klimazwiebel comes in. He initiated a valuable, open discusssion about it, including your contributions, and that what he is to be applauded for. -- Hans von Storch

eduardo said...

Jesús, gracias por tus comentarios

First, a minor point. Are you perhaps confusing 'global mean' with 'most likely value'? They are very different terms.

If it is fair to accept an informal comment by one of the authors assessing his own work, without hearing what Siddal et al. have to say, it is also fair to state that Rahmstorf 2007 has been demonstrated to be wrong by two peer-reviewed comments published in Science, right?
Well, allow me to say that the authors of realclimate and the authors of the CD are closely related, so I cannot consider a comment in realclimate as an independent assessment. I think we should wait until a comment on Siddal et al. and a response are published in a scientific journal. If you are by then right, I would be happy to give you this point. Until then, to ignore Siddal et al. is cherry picking.

Let us consider Pfeffer et al. The CD probably quotes it in the headlines as :
'Accounting for ice-sheet mass loss, sea-level rise until 2100 is likely to be at least twice as large as that presented by IPCC AR4, with an upper limit of ~2m based on new ice-sheet understanding'

whereas the real Pffeffer et al abstract reads:

'We consider glaciological conditions required for large sea-level rise to occur by 2100 and conclude that increases in excess of 2 meters are physically untenable. We find that a total sea-level rise of about 2 meters by 2100 could occur under physically possible glaciological conditions but only if all variables are quickly accelerated to extremely high limits. More plausible but still accelerated conditions lead to total sea-level rise by 2100 of about 0.8 meter.'

To me both sound very different. Would you think that the CD headline is a fair summary?

Another paper, Katsmann et al, 2008, about North Atlantic sea-level, (note that '..the analyzed AOGCM simulations, sea level in the North Atlantic Ocean increases more than the global mean sea level.')

The resulting set of climate scenarios represents our best estimate of twenty-first century sea level rise in the northeast Atlantic Ocean, given the current understanding of the various contributions. For 2100, they yield a local rise of 30 to 55 cm and 40 to 80 cm for the moderate and large rise in global mean atmospheric temperature, respectively.

This also sounds very different to '... as much as 2 meters'.

The CD has turned to be irrelevant in the end. Perhaps the reason is that it was perceived as biased? Let me explain what I mean. I would have rephrased the headline summary as following : 'Future sea-level rise projections for 2100 indicate a most likely value of about 80 cm, 4 times the rise in the 20th century. A sea level rise of 2 meters could occur under extreme scenarios.'

Thus a policy maker gets a better summary about the risks and chances, and more importantly, it is percieved as more credible. Make no mistake, 80 cm of sea-level rise is already bad enough.

Anonymous said...

Hans von Storch #27,

Thanks for your comments and your tone.

I don’t think that many scientists go along with R07 because of R’s own opinion or because of the CD (which is later than the papers I cited). I think that scientists make their own opinion and, after reading the critics and R’s response, they conclude that there is no fundamental problem with R07.

I find really striking that you compare a pioneering breakthrough such as R07 or MBH98 (or their revision process) with a blatantly wrong paper such as Soon & Baliunas 03. Equaling repair (SB03) and improvement (R07, MBH98) is like equaling wrong (something in need of repair) and right (something that can be improved). Whereas SB03 just brought more confusion and meant a step backward, R07 or MBH98 brought a better understanding than before; a step forward; something indeed to be applauded for. -- Jesús Rosino.

Anonymous said...

Eduardo #29,

The alternative headline you’ve suggested is interesting, as we would more or less agree about the facts, but may disagree on how to communicate them. However, that's not what people are reading in this blog post. People are reading that we are still considering a maximum 21st century SLR of 82 cm and that most studies conclude that sea level cannot possibly rise more than 1 meter. The CD does a much better job in summarizing the advances after the IPCC AR4.

The CD is not a comprehensive assessment, it’s just a quick update; its purpose is not to review all the peer reviewed literature, but to give a quick overview of the main progress. AFAIK, Soon and Baliunas did not publish any public correction, nor did McKitrick & Michaels. Would you include those papers in a quick summary of the climate science advances? Would you include Andreae et al 2005 or Stainforth et al 2005 when summarizing the upper limit of climate sensitivity? Being published in a peer reviewed journal doesn’t guarantee equal merits and not every error is publicly corrected. Is your assessment of the science that Rahsmtorf was convincingly refuted whereas Siddal et al got the science right? Quite weird, since Siddal used that same method as Rahmstorf that was considered wrong. What I really think is that this would be just a formal (and thus meaningless) assessment.

As for Milne et al, I still maintain that they don’t rule out a SLR of more than 1 m. That would be a wrong assessment of the post-IPCC papers and they explicitly stated it until the very last moment. Still, I must acknowledge that it’s entirely my fault that I trusted a translation of your blog post that omitted the word “likely” and transformed your words “will likely remain under 1m” into “will remain under 1m” [;msg2227985#msg2227985]. That said, the statement that SLR will *likely* remain under 1m doesn’t even contradict R07. And it is still double the IPCC AR4’s numbers (0,18 - 0,59 m), as stated in the CD. "Upper limit of 2 m" or "2 m. under extreme scenarios" are both ok for me.

As for Katsman et al 2008, that’s not really a new approach compared to the IPCC AR4, and we have good reasons to believe that physical models underestimate SLR.

Your proposed headline, on the other hand, is focused on Pfeffer et al. How about including Grinsted et al?: “Sea level 2090–2099 is projected to be 0.9 to 1.3 m for the A1B scenario, with low probability of the rise being within IPCC confidence limits”. Besides, A1B cannot currently be considered a realistic scenario, we are following A1FI instead, for which they project 1.1 to 1.6 meters.

Anyway, I think that this last message from you is a more informative, balanced and interesting view.

Gracias por vuestros comentarios y vuestra actitud. Un saludo,

Jesús Rosino

TCO said...

I have read MBH and while it was ambitious, don't think it was groundbreaking or pioneering. The methodolgy was so hosed and so poorly disclosed, that I don't think it can lay credit to that sort of stake. Remember if you are going to say a word that you have read, but don't know how to pronounce, SAY IT LOUD. Mann mumbled on the methodology. His writing is a case of trying to act fancy to cloak things, rather than like a real scientist to show and disclose.

Jesús Rosino said...

TCO #31,

Well, in the US NAS own words, MBH was "the first systematic, statistically based synthesis of multiple climate proxies" and "the first to include explicit statistical error bars, which provide an indication of the confidence that can be placed in the results". That's pioneering for me. According to Wahl & Ammann 05, "the MBH reconstruction, per se, is closely reproducible if the procedural sequence of MBH is followed and all original proxy data are applied".

eduardo said...

This Times article is related to the sea-level controversy. Note that Holgate, Gregory and Rahmstorf are well known IPCC authors

Jesús Rosino said...

#23 Eduardo said: "I would like to see a public correction of the Siddal et al paper. A comment by Eric Stieg in realclimate starting with ' as far as I am aware...' is really not enough. Realclimate has unfortunately not the best record of accuracy and unbiasedness"

Although it was unnecessary (as it was crystal clear after the thorough explanation published by Rahmstorf and Vermeer in RC), the authors have took the trouble to specifically retract their paper:

Would you please retract your comments based on this obviously flawed paper? Thanks

As mentioned over and over again, peer review is a necessary but not sufficient condition, or, in other words, substantial comments (as RC's on that paper) has much more credibility than merely formal comments ("it is in the peer reviewed literature, so it's sacred unless the authors acknowledge their error"). The latter just breeds the never-ending conspiracy theories from denialists.

eduardo said...

Hola Jesus,

yes, Siddal et al retracted their paper and they have to be congratulated by their honesty. So I have no problem to say that this paper was not correct.

I never said that peer review is a sufficient condition or that a published paper is sacred.. Actually, many published papers turn to be wrong. I do not understand why you say this retraction was unnecessary. It was indeed necessary that the authors themselves explain which was in error in the paper. I have another opinion from the comments published in realclimate than you may have. The fact that realclimate applies a strong filtering on the comments posted by others makes the discussion of scientific papers there unreliable.

What the Copenhagen Diagnosis did was to simply ignore this paper and the review by Milne, without explaining why.. For instance, I still think that Rahmstorf 2007 is wrong, but I always have the chance to publish my results and offer other scientist the possibility to correct me. What I would not do is to trash Rahmstorf paper here and block all his possible comments. Comments in a blog in which the other side has not the opportunity to react are worthless- this applies to all blogs with that policy, not only to realclimate. So it was not crystal clear at all.

Now if you are curious and have the chance, please comment with a statistician of your choice the Rahmstorf paper and the two comments to it that were published in Science. You may be surprised.

Anyway, thanks for commenting here

Georg said...

Hola Edu

"What I would not do is to trash Rahmstorf paper here and block all his possible comments. "

I would be very interested to have you commenting on Rahmstorfs sea level reconstructions. He certainly could comment here and could even be invited to do so by an email. I dont know if this will be a "trashing" but critical comments and discussion are obviously no problem.

What this post is until now is again a metadiscussion. Were arguments/papers legally used the way the should be used etc. But science is not only about "correct use" of information but also on "information" itself, right.

The only interesting claim made here is that Stefan Rahmstorf is completely ignorant about moving coast lines (see above by Richard Tol). And frankly I have problems believing him. In any case why not an article on the statistical issues you were mentioning?

Saludos Georg

Jesús Rosino said...

Eduardo (#35), let me sum up how I see this conversation on Siddal et al:

Edu: why these papers weren't cited? We can rule out the quality of the papers being problematic because they were peer reviewed. So they aren't cited because Rahmsotrf doesn't like the conclusion.

Me: Siddal wasn't cited because they considered it to be flawed, as explained in RC.

*The next logical step would have been going on to discuss the analysis set out at RC. Otherwise, the conversation was dead there.

However I added some extra information: the authors acknowledged it (according to Eric Steig).

And here is where I think that the conversation drifted toward the metadiscussion mentioned by Georg's:
Edu: I don't trust Eric, the paper is right unless the authors retract it.

Me: the authors have retracted it,
even though it was not necessary [it is not necessary for Rahmstorf or any other expert in order to assess the errors on it; a lot of wrong papers are not corrected, as I said before; I don't need Soon and Baliunas retracting their paper to know it is wrong; I don't need Lindzen retracting his Iris paper to know it is unsubstantiated; I don't need Svensmark retracting his assertion that cosmic rays play a big part in the everyday weather to know he has not proven that]. A public retraction is just desirable, but it is not required to assess the errors of a work. My point is that it is now clear that Rahmstorf was right when assessing the Siddal et al paper, and the fact that Siddal et al decided to acknowledge it or not doesn't make any difference to that. If you wanted to rehabilitate Siddal et al, I think your only option would have been to go into the analysis provided by Rahmstorf and Vermeer in RC and explain us why they were wrong. That would have been very interesting.


eduardo said...



there are a couple of points: the Copenhagen Diagnosis was meant to be an assessment of the science for policy makers. It was not meant to be a blog posting. That is the critical difference. A policy maker reading an assessment needs to know why papers are ignored in an assessment.

You think that a posting in realclimate is explanation enough. Apart from the question of why should a policy maker read realclimate among the thousands of other climate blogs, I dont think it is enough, because debates in realclimate are censored. I could not have posted in realclimate because my comments are not published there. werent you aware of that? how many other comments are excluded? I cannot know. So I have not reason to trust realclimate, even less in this case in which they were assessing a paper that contradicted one of its own. By the same token I would not take for granted a rebuttal of any paper in climate audit, unless I have the opinion of the authors.

In other cases, in which realclimate also asserted that papers were wrong, turned out that realclimate was wrong . For instance, von Storch et al.2004, which was confirmed by later analysis, even in the NAS report.

Other than that I envy you for being able to assess papers without a shade of doubt in so different areas. In a blog, that may suffice. If you want to convince others in an assessment, it may not

Jesús Rosino said...

Eduardo: "A policy maker reading an assessment needs to know why papers are ignored in an assessment."

No, he/she doesn't.

Eduardo: "You think that a posting in realclimate is explanation enough."

I think that the specialist is qualified to assess what's NOT worth mentioning without any need to explain his/her reasons anywhere. Of course, if the information is required because of social pressure, any mean is equially valid, even a private e-mail. Making it public or publishing it is, in fact, a decision about politics, not science; it is a decision for a policy maker.

eduardo said...

Hola Jesus,

ok, then do you think it would be perfectly right if I write an assessment about future sea-level and I do not mention Rahmstorf's papers, or about climate reconstructions and I ignore Mann's papers.

Well, no. I would mention them and if I do not agree with their content I would at least point to a source where an explanation of my reasons can be found. This is by the way usual when reviewing a manuscript: to assure that the relevant literature is mentioned and discussed. If I were a policy maker and my scientific adviser does not make me aware of all relevant results - I would certainly not trust him/her any more.

I think your position is quite non-scientific. You assume that there is one specialist that holds the truth - a gatekeeper in other words. This is exactly the opposite of what science is.

Hans von Storch said...

Jesus Rosino - I guess you accept me as a specialist, right? Good. Then, you will be satisfied with my assessment that your opinion, which you ave voiced above (39) is not acceptable. Using your arguments I do not need to give reasons for my assessment - because I am a specialist in the field.
But you will not mind that I will in future use your quote as an example, of what type of weird ideas some people have.