Will Replicated Global Warming Science Make Mann Go Ape?
By Patrick J. Michaels – from World Climate Report
About 10 years ago, December 20, 2002 to be exact, we published a paper titled “Revised 21st century temperature projections” in the journal Climate Research. We concluded:
Temperature projections for the 21st century made in the Third Assessment Report (TAR) of the United Nations Intergovernmental Panel on Climate Change (IPCC) indicate a rise of 1.4 to 5.8°C for 1990–2100. However, several independent lines of evidence suggest that the projections at the upper end of this range are not well supported…. The constancy of these somewhat independent results encourages us to conclude that 21st century warming will be modest and near the low end of the IPCC TAR projections.
We examined several different avenues of determining the likely amount of global warming to come over the 21st century. One was an adjustment to climate models based on (then) new research appearing in the peer-reviewed journals that related to the strength of the carbon cycle feedbacks (less than previously determined), the warming effect of black carbon aerosols (greater than previously determined), and the magnitude of the climate sensitivity (lower than previous estimates). Another was an adjustment (downward) to the rate of the future build-up of atmospheric carbon dioxide that was guided by the character of the observed atmospheric CO2 increase (which had flattened out during the previous 25 years). And our third estimate of future warming was the most comprehensive, as it used the observed character of global temperature increase—an integrator of all processes acting upon it—to guide an adjustment to the temperature projections produced by a collection of climate models. All three avenues that we pursued led to somewhat similar estimates for the end-of- the-century temperature rise. Here is how we described our findings in paper’s
Abstract:
Since the publication of the TAR, several findings have appeared in the scientific literature that challenge many of the assumptions that generated the TAR temperature range. Incorporating new findings on the radiative forcing of black carbon (BC) aerosols, the magnitude of the climate sensitivity, and the strength of the climate/carbon cycle feedbacks into a simple upwelling diffusion/energy balance model similar to the one that was used in the TAR, we find that the range of projected warming for the 1990–2100 period is reduced to 1.1–2.8°C. When we adjust the TAR emissions scenarios to include an atmospheric CO2 pathway that is based upon observed CO2 increases during the past 25 yr, we find a warming range of 1.5–2.6°C prior to the adjustments for the new findings. Factoring in these findings along with the adjusted CO2 pathway reduces the range to 1.0–1.6°C. And thirdly, a simple empirical adjustment to the average of a large family of models, based upon observed changes in temperature, yields a warming range of 1.3–3.0°C, with a central value of 1.9°C.
We thus concluded:
Our adjustments of the projected temperature trends for the 21st century all produce warming trends that cluster in the lower portion of the IPCC TAR range. Together, they result in a range of warming from 1990 to 2100 of 1.0 to 3.0°C, with a central value that averages 1.8°C across our analyses.
Little did we know at the time, but behind the scenes, our paper, the review process that resulted in its publication, the editor in charge of our submission, and the journal itself, were being derided by the sleazy crowd that revealed themselves in the notorious “Climategate” emails, first released in November, 2009. In fact, the publication of our paper was to serve as one of the central pillars that this goon squad used to attack on the integrity of the journal Climate Research and one of its editors, Chris de Freitas.
The initial complaint about our paper was raised back in 2003 shortly after its publication by Tom Wigley, of the US National Center for Atmospheric Research and University of Toronto’s L. D. Danny Harvey, who served as supposedly “anonymous” reviewers of the paper and who apparently had a less than favorable opinion about our work that they weren’t shy about spreading around. According to Australian climate scientist Barrie Pittock:
I heard second hand that Tom Wigley was very annoyed about a paper which gave very low projections of future warmings (I forget which paper, but it was in a recent issue [of Climate Research]) got through despite strong criticism from him as a reviewer.
So much for being anonymous.
The nature of Wigley and Harvey’s dissatisfaction was later made clear in a letter they sent to Chris de Freitas (the editor at Climate Research who oversaw our submission) and demanded to know the details of the review process that led to the publication of our paper over their recommendation for its rejection. Here is an excerpt from that letter:
Your decision that a paper judged totally unacceptable for publication should not require re-review is unprecedented in our experience. We therefore request that you forward to us copies of the authors responses to our criticisms, together with: (1) your reason for not sending these responses or the revised manuscript to us; (2) an explanation for your judgment that the revised paper should be published in the absence of our re-review; and (3) your reason for failing to follow accepted editorial procedures.
Wigley asked Harvey to distribute a copy of their letter of inquiry/complaint to a large number of individuals who were organizing some type of punitive action against Climate Research for publishing what they considered to be “bad” papers. Apparently, Dr. de Freitas responded to Wigley and Harvey’s demands with the following perfectly reasonable explanation:
The [Michaels et al. manuscript] was reviewed initially by five referees. … The other three referees, all reputable atmospheric scientists, agreed it should be published subject to minor revision. Even then I used a sixth person to help me decide. I took his advice and that of the three other referees and sent the [manuscript] back for revision. It was later accepted for publication. The refereeing process was more rigorous than usual.
This did little to appease to those wanting to discredit Climate Research (and prevent the publication of “skeptic” research) as evidenced by this email from Mike Mann to Tom Wigley and a long list of other influential climate scientists:
Dear Tom et al,
Thanks for comments–I see we’ve built up an impressive distribution list here!
…
Much like a server which has been compromised as a launching point for computer viruses, I fear that “Climate Research” has become a hopelessly compromised vehicle in the skeptics’ (can we find a better word?) disinformation campaign, and some of the discussion that I’ve seen (e.g. a potential threat of mass resignation among the legitimate members of the CR editorial board) seems, in my opinion, to have some potential merit.
This should be justified not on the basis of the publication of science we may not like of course, but based on the evidence (e.g. as provided by Tom and Danny Harvey and I’m sure there is much more) that a legitimate peer-review process has not been followed by at least one particular editor.
Mann went on to add “it was easy to make sure that the worst papers, perhaps including certain ones Tom refers to, didn’t see the light of the day at J. Climate.” This was because Mann was serving as an editor of the Journal of Climate and was indicating that he could control the content of accepted papers. But since Climate Research was beyond their direct control, it required a different route to content control. Thus pressure was brought to bear on the editors as well as on the publisher of the journal. And, they were willing to make things personal. For a more complete telling of the type and timeline of the pressure brought upon Chris de Freitas and Climate Research see this story put together from the Climategate emails by Anthony Watts over at Watts Up With That.
Now, let’s turn the wheels of time ahead 10 years, to January 10, 2012. Just published in the journal Geophysical Research Letters is a paper with this provocative title: “Improved constraints in 21st century warming derived using 160 years of temperature observations” by Nathan Gillet and colleagues from the Canadian Centre for Climate Modelling and Analysis of Environment Canada (not a group that anyone would confuse with the usual skeptics). An excerpt from the paper’s abstract provides the gist of the analysis:
Projections of 21st century warming may be derived by using regression-based methods to scale a model’s projected warming up or down according to whether it under- or over-predicts the response to anthropogenic forcings over the historical period. Here we apply such a method using near surface air temperature observations over the 1851–2010 period, historical simulations of the response to changing greenhouse gases, aerosols and natural forcings, and simulations of future climate change under the Representative Concentration Pathways from the second generation Canadian Earth System Model (CanESM2).
Or, to put it another way, Gillet et al. used the observed character of global temperature increase—an integrator of all processes acting upon it—to guide an adjustment to the temperature projections produced by a climate model. Sounds familiar!!
And what did they find? From the Abstract of Gillet et al.:
Our analysis also leads to a relatively low and tightly-constrained estimate of Transient Climate Response of 1.3–1.8°C, and relatively low projections of 21st-century warming under the Representative Concentration Pathways.
The Transient Climate Response is the temperature rise at the time of the doubling of the atmospheric CO2 concentration, which will most likely occur sometime in the latter decades of this century. Which means that results of Gillet et al. are in direct accordance with the results of Michaels et al. published 10 years prior and which played a central role in precipitating the wrath of the Climategate scientists upon us, Chris de Freitas and Climate Research.
Both the Gillet et al. (2012) and the Michaels et al. (2002) studies show that climate models are over-predicting the amount of warming that is a result of human changes to the constituents of the atmosphere, and that when they are constrained to conform to actual observations of the earth’s temperature progression, the models project much less future warming (Figure 1).
Figure 1. Dashed lines show the projected course of 21st century global temperature rise as projected by the latest version (CanESM2) of the Canadian coupled ocean‐atmosphere climate model for three different future emission scenarios (RCPs). Colored bars represent the range of model projections when constrained by past 160 years of observations. All uncertainty ranges are 5–95%. (figure adapted from Gillet et al., 2012: note the original figure included additional data not relevant to this discussion).
And a final word of advice to whoever was the editor at GRL that was responsible for overseeing the Gillet et al. publication—watch your back.
References:
Gillet, N.P., et al., 2012. Improved constraints on 21st-century warming derived using 160 years of temperature observations. Geophysical Research Letters, 39, L01704, doi:10.1029/2011GL050226.
Michaels, P.J., et al., 2002. Revised 21st century temperature projections. Climate Research, 23, 1-9.
Dana,
If you think “the entire post is utter nonsense” (though I don’t believe so), then by the same reasoning, the entire work by Mann and the “team” is utter nonsense. The emails reveal that they’re not trustworthy.
Who died made you dictator? You made your bed, now sleep in it.
I should also point out that the Gillett results highlighted by Michaels/Chip are based on the Gillett estimate that the planet warmed 0.7°C from 1961 to 2010, vs. just 0.6°C from 1851 to 2010. That’s why he gets a lower transient response using the longer timeframe. That’s why it’s important to also include the 1901-2000 regression results (because frankly they underestimate the 1851-2010 warming, which biases their sensitivity and future projections low). If you ignore the latter and focus entirely on the 1851-2010 results, you’re accepting that the planet has warmed 0.7°C since 1961, and that it cooled from 1851 to 1961.
This would of course contradict the surface stations ‘urban heat island’ arguments put forth so often on this site. Whoops.
dana1981,
Thanks for your comments.
Obviously, there are more details, caveats, etc. contained in the full version of the paper than in our article which focused on the primary result. If these additional caveats, details, etc. overwhelmed the main results, then the paper would not have been published in the first place, so I think it is fair of us to place the focus where we did.
Just because the authors chose to show some combination of data in their Figure does not obligate us to show the same combination of data. Perhaps their intensions for their plot were different than ours. Again, for our purposes, we were interested in the main, new, result from Gillett et al., not how the new result differed from a result using older data. While that is perhaps an interesting academic result, it just wasn’t what our article was about.
-Chip Knappenberger
World Climate Report
I just posted this link to the (now) Senior Executive Producer of NOVA, who guided the series
for a couple of decades.
But, I agree with:
(Completely + Hopelessly)Politicised = (Fat)Chance
Maybe Americans for Prosperity would chip in? David Koch is an MIT grad….
…Lady in Red
It`s great that dana1981`s comment were allowed to go through. This is what makes this blog so relevant. She raises a good point, allowing for a counter-response. At the end, I feel I`m a little less stupid… Thank you!
[Moderator’s Note: Dana is a he. -REP]
mkelly said:
“Mr. Gates your quote uses the term “doubling of CO2″ and you also use 560 as a number, so 280 must have been half to accomplish a doubling. Going from 1 to 2 is doubling. So per your statement a doubling of CO2 from 1 to 2 caused an increase of 1.5C. ln(2/1)= ln(560/280)
So I ask again does it give you pause to say a doubling of CO2 will lead to a sensitivity of 1.5C?”
______
I’m really glad you brought this up as it offers a chance to really have a discussion about the issue of the logarithmic effect of CO2 and sensitiy. First, you absolutely cannot put climate sensitvity to varioius amount of CO2 on a log chart, as much as skeptics would love to do this. It is absolutely an incorrect thing to do. The logarithmic response of concentrations of CO2 tells you nothing about the climate sensiviity to those concentrations. Once honest skeptics understand this (I mean really understand it) they stop doing it, and move on to other issues. The logarithmic response to increasing amounts of CO2 is a completely separate issue to climate sensitivity, which involves the response of the climate system to a given concentraion of CO2 in the atmopshere. The climate sensitivity involves not just the logarithmic radiational response of CO2, but all additional feedbacks, both positive and negative to that radiational response. But if gets even worse for those continuing on in their “but it’s logarithmic” theme. There are mutliple levels of climate sensitivity, such that you need to clearly define what kind of sensitivity you are talking about. The main general two are Transient (also called Charney) and Earth System. Transient sensitivity takes into account all immediate and short-term feedbacks to a given level of CO2. Note: Basic Transient or Charney sensitivity to given CO2 concentrations is not at all logarithmic, but rather involves the complexity of the system itself, such that senstivity to any given level is not projectable on a simply log chart as you can’t project the response of a complex dynamical system existing on the edge of chaos on a log chart! It is simply incorrect. The reason being is that at different levels of concentrations a chaotic system such as the climate will jump to an entirely new regime, not predictable from following a simple log chart. But here’s the even worse part. The final Earth system or Equilbrium sensitivity doesn’t happen right away, but can take decades to finally settle out due to longer term slow feedbacks. If it was wrong to put transient sensitivty on a log chart, it is infinitely even more incorrect to put equilibrium sensitivt there, and Lord Monckton’s proclamation that they are “nearly equal” is absolutely not founded on anything scientific, as we don’t even know for certain what the transient sensitivity is, how can we possibly know that the even much more difficult to know, equalibrium sensitivity is “nearly equa”. You can’t, and Lord Monckton can’t. But what you especially can’t do, is put either transient or equilibrium sensitiity on a log chart.
dana1981
“To delete results from the figure just because they weren’t highlighted (but were discussed) in the paper is a misrepresentation and distortion. The paper also had a number of other caveats, for example”
Kinda like hide the decline.
the defense of hide the decline was that the decline wasnt hidden because
1. it was shown elsewhere
2. it was discussed elsewhere
So, if you apply that defense here dana..
Or rather, if you insist that deleting results here is misleading, then you also must condemn mikes nature trick and hide the decline
Looks like dana1981 has been caught out. He/she barged in with the claim that Pat and Chip were dishonestly hiding the half of Gillett’s paper that contradict their interpretation. Chip’s explanation shows that the omitted results (which the posts informs us are being omitted as not to the point) simply reinforce their interpretation of the paper. The omitted data shows how much Gillett’s estimated climate sensitivity range lowers when models are fit to the longer temperature record (bringing it close to what Dr. Michaels estimated 10 years ago).
Now dana is left sputtering about how he/she prefers the shorter temperature record (you know, because throwing away information is the new “scientific method”). But the suggestion that Gillett will agree with her is ludicrous. Did dana not see the title of Gillett’s paper: “Improved constraints in 21st century warming derived using 160 years of temperature observations”? IMPROVED. That’s the key word. So keep on digging diva dana.
It will be richly ironic that if the down fall of the ‘the Team’ comes about in the end not becasue ‘science’ in general calls out their poor behavior, but that their arrogance leads them to stuff themselves up in the eyes of the public .
dana1981 and yet you can’t apply the same logic to a person , in one instance , that not only used data he knew to be rubbish but actual used in upside too, such is the ‘quality ‘ of Mann’s work.
But then again anything in the name of ‘the cause ‘ most it good in the eyes of the faithful.
Dana.
“based on the 1901-2000 regression are more accurate, though Gillette et al. chose to focus on the results based on the 1851-2010 regression”
OF COURSE the warmist bletheren would prefer an analysis that leaves out the last 10 years or so.
“This post is a dishonest disgrace. The figure deletes essentially half of the Gillett (which is the correct spelling, by the way – two ‘t’s) results, calling them “not relevant”. More like not convenient, because they show Gillett’s second possible climate response, which is one with significantly higher warming. The full figure can be seen here”
Yup, kinda like hiding the decline dana.
When SkS defended hiding the decline, they appealed to two arguments.
1. it was discussed in the text
2. it was shown in other texts.
The WMO hide the decline was defended on the basis that the publication was not as important as IPCC.
It appears that
1. this blog post is not as important as the IPCC
2. the fact that the graphic has been changed is mentioned in the legend
3. the graphic is shown in other places.
using the defenses you used for hide the decline, this graph passes muster
On the other hand, if you call this graph misleading and raise a stink about a blog post,
then you should reconsider your defense of hide the decline. because here you are holding a blog to higher standard than the IPCC
Climate research and medical/pharmaceutical research are very different animals. It is amusing for me to watch. Reading about climatology and climastrology has been a hobby of mine for the last 7 years or so. Keeping up with the medical literature is a part of my livelihood. The statistical analyses employed in medical/pharmaceutical research are generally pretty straightforward. They’re not obscure or arcane and can usually be readily understood by those with a fundamental understanding of statistics. The statistical analyses employed in climate science make my eyes glaze over.
There are all kinds of other differences. The fields of medicine and pharmacology are much, much larger than climatology. Most medical and pharmaceutical research is funded by the private sector and is subjected to intense review and analysis both before and after publication. Apparently this doesn’t happen in the field of climatology. Then again, nobody dies if climatologists screw up.
About 25 years ago some of the pharmaceutical industries played a nasty little game. They would fund university based research (which were often Phase III clinical trials or post-marketing studies). This ain’t cheap. Some pharmaceutical companies forked over the research funds but stipulated that they had the final decision if the study would be published. In essence they were a position to stop publication of results they didn’t like. What’s the point of conducting research if you cannot communicate the results? In very short order the ship hit the sand. Grant applications were restructured so that all the pharmaceutical industry could do is write a check. The researchers were free to publish whatever results they got. In medicine negative results are just as interesting as positive results. Then again, their funding doesn’t depend on achieving results that serve a political agenda.
I’ve served as a reviewer before. I can corroborate the rumor that it is a tedious chore and basically a pain in the ass. But I was truly dispassionate. I didn’t care about the results, my job was to examine the study design and analysis. Beyond that I didn’t care. The rest was up to the journal editor. Climate science seems to be quite different. The results are the most important factor, the validity of the study design and subsequent analysis are secondary.
I don’t mean to give the impression that I glorify medical research. The manner in which they report results are still very much in accordance with the desires of the funding entity. It’s just harder to tell bald faced lies. I have several examples but I won’t bore WUWT readers with more than just one. A rather famous study found that using (at the time proprietary) clopidogrel at $3 or more per day reduced the risk of subsequent stroke by 50% over using aspirin (at about 2 cents per day). This is absolutely true. What they prefer not to mention is that they’re looking at a relative risk difference of 1.6% vs. 0.8%….at a cost of over $1,000 per year per patient. To MAYBE prevent one subsequent stroke event a huge number of patients have to be treated with the expensive drugs for several years. Is is statistically significant? Yeah. Is it clinically significant? Not really. Is it worth it? Nope.
Chip, I agree that your intentions are clearly different. Gillett et al. intended to show that their results depend on the timeframe used in the regression. You and Michaels intended to leave out the data that undermine your argument. I already explained that your claim about ‘new’ and ‘old’ data is utterly nonsensical. The difference between the two results is merely the difference between using the regressions over different timeframes, as the paper clearly discusses. To delete the 1901-2000 regression is simply to delete the results you don’t like. And it is quite obvious why you don’t like them.
Mosher – ‘hide the decline’ and ‘Mike’s Nature trick’ aren’t even the same thing. I suggest you research the subject before commenting on it. No, the two examples are not even remotely similar.
Yes, I am a ‘he’, thank you for clarifying, moderator.
Alec Rawls says:
January 11, 2012 at 12:17 pm
Absolutely! I suspect Dana1981 is too young (31?) or inexperienced to understand that care in thought is required before writing. Chips’ explanation makes sense to me too.
R Gates – “The equalibrium sensitivity will be higher, but exactly how much higher, no one knows, as a full understanding of all the earth-system feedbacks is no known.”
The equilibrium sensitivity to doubling CO2 is precisely zero.
CO2* + N2 CO2 + N2⁺
In equilibrium, CO2 does not heat the atmosphere.
The atmosphere is heated via convection/conduction from the surface.
Kinda reminds me of Svensmark.
Seconded. I suspect it’s part of the evidence that ought to come out. Thank you for visiting, Dr Michaels.
Hmm the equation in my last post needs ↔, greater than, less than symbols.
Pat Michaels – any chance you can email me on the address you last used ?
Thanks.
Warwick Hughes
The Abstract of the Gillett et al article is actually more clearly in support of Michaels’ points than are the portions of it Michaels quotes above. Here is what the full Abstract says, using all caps to isolate the part the Michaels left out, and which demolishes what Dana has suggested is bias on Michaels’ part:
“Projections of 21st century warming may be derived by using regression-based methods to scale a model’s projected warming up or down according to whether it under- or over-predicts the response to anthropogenic forcings over the historical period. Here we apply such a method using near surface air temperature observations over the 1851–2010 period, historical simulations of the response to changing greenhouse gases, aerosols and natural forcings, and simulations of future climate change under the Representative Concentration Pathways from the second generation Canadian Earth System Model (CanESM2). Consistent with previous studies, we detect the influence of greenhouse gases, aerosols and natural forcings in the observed temperature record. OUR ESTIMATE OF GREENHOUSE-GAS-ATTRIBUTABLE WARMING IS LOWER THAN THAT DERIVED USING ONLY 1900-1999 OBSERVATIONS. Our analysis also leads to a relatively low and tightly-constrained estimate of Transient Climate Response of 1.3–1.8°C, and relatively low projections of 21st-century warming under the Representative Concentration Pathways. Repeating our attribution analysis with a second model (CNRM-CM5) gives consistent results, albeit with somewhat larger uncertainties.”
Dana, when the authors themselves say, in the abstract summarizing the article, that “Our estimate of greenhouse-gas-attributable warming is lower than that derived using only 1900–1999 observations,” then the authors mean for readers to understand that the longer observations are preferable. When they use the diminutive term “only,” they mean to say that the shorter time period is insufficient.
Therefore, when you “alert” Gillett to what Michaels has done, I doubt very much he will think Michaels has distorted anything, or taken anything out of context. Please do “alert” Gillett, and make sure to report back in full.
Until now, I had not known how intellectually dishonest you are, Dana. You are trying to trick people who haven’t actually read the full Abstract into thinking that Michaels is hiding from WUWT readership that Gillett et al actually came to different conclusions than Michaels says they do. Nice try.
dana1981,
I think it is pretty clear which timeframe that Gillet et al. preferred in their article “Improved constraints in 21st century warming derived using 160 years of temperature observations.”
-Chip Knappenberger
@ur momisugly John
“they mean to say that the shorter time period is insufficient. ”
As is nearly always the case.
Now if we had decent accurate data to add in the MWP and RWP, then I suggest that the result would show even less sensitivity to CO2.. like nearly ZERO !!!!
Unfortunately we have to rely on only the very small time span of 150 years of data (approx), which means that predictions are probably limited to 5-10 years, even then with large errors expected.
It really is time for the warmist bletheren to realise this and stop making moronic predictions for the end of the century.
Genghis says:
January 11, 2012 at 12:49 pm
R Gates – “The equalibrium sensitivity will be higher, but exactly how much higher, no one knows, as a full understanding of all the earth-system feedbacks is no known.”
The equilibrium sensitivity to doubling CO2 is precisely zero.
CO2* + N2 CO2 + N2⁺
In equilibrium, CO2 does not heat the atmosphere.
The atmosphere is heated via convection/conduction from the surface.
_______
Whatever book, pamplet, or website you got this from, recycle, discard, and unlink, and try taking a real physics class.
John, I read it the same way. The “Key Points” further express preference for the 1851-2010 time period as “Using 1851-2010 observations gives lower AND LESS UNCERTAIN projected warming”
dana1981 your outrage and hysteria seem uncorrelated to and out of scale with this article. It is confusing.
Lady in Red:
I have just submitted feedback to NOVA essentially trying to do what I suggested earlier.
If you have an email address you can share, I authorize WUWT to release my email addr to you so that you can contact me and share a better way to get the message driven home.
Getting NOVA to program this story, without leftist distortions, should be the highest priority work, in my opinion. After the years of pushing the agenda, it is PAST time that NOVA start practicing a little balance.
thanks for your efforts, Steve
[DONE. -REP]