Climate Audit has been considered – at least as a phenomenon – in a couple of courses. Kenneth Blumenfeld’s students had a different reaction than the Georgia Tech students. Earlier this year, I wrote a short comment about a post that Kenneth had made at realclimate about this, which I am re-posting in its entirety. It’s amazing how perceptions differ. The students had to represent various viewpoints and one even chose to represent CA. Kenneth recently asked me to write another into piece of a similar nature for the fall term which I’ve appended below – following the reprint.
Our "blogfather", realclimate, has been celebrating their one-year anniversary (congratulations to them) and have been reflecting on their year. Kenneth Blumenfeld, who’s posted here once or twice, posted an interesting comment at realclimate, about how his undergraduates were now investigating climate change issues online, that they "very badly wanted to get behind RC", but wanted them to "step up to the plate, not just take practice swings", mentioning that they were getting their butts kicked. To any such undergraduates that may have come to this site: welcome.
Kenneth Blumenfeld’s full comment is here as follows:
I should relay, however, that undergraduate meteorology, geography, and geology students (to name only few disciplines) are now taking time out to investigate these issues, and they are doing it online, rather than in the literature, as wacko as that may seem. Despite the overwhelming majority of consensus-side scientists out there, it is much easier to get contrarian/skeptical/psuedo skeptical information. The consensus folks are getting their you-know-whats kicked in this regard, and all I can offer as evidence are my 90 or so students this past semester who I think very badly wanted to get behind RC but felt they were side-stepping direct confrontation. They felt that by appearing to ignore skeptics (except for on its own forum) RC was in some sort of denial. You and I may not believe this is true, but to the future climate scientists I think it is an important point. They want to see their people step up to the plate, not just take practice swings, so to speak.
Gavin’s reply is quite revealing of realclimate attitudes towards laity:
I think one needs to differentiate dealing with ‘sceptic’ issues from going head-to-head with some particular site or person. … [I] think it is do what we are doing – provide solid discussions of the real scientific issues which can then be used by others in different forums. If you have any specific ideas to make that work better, let us know.
A few thoughts for such undergraduates (to regulars, I apologize for repeating some old stories):
As I often repeat, I am not a "contrarian". If I were a politician and forced to make a decision on climate policy in the next 10 minutes, I would be guided by the IPCC and the various learned societies that I so often criticize. However, any scientist worth his salt (as Feynmann tells us) should not rely on authority and should question authorities. Such inquiries at realclimate often provoke a highly irritating faux exhaustion ("…sigh,…"). Not here.
Anyway, like an inquiring student, I’ve taken an interest in questions of climate change, with a view to understanding exactly how IPCC climate scientists were able to come to the conclusions that inform their policy recommendations. The most prominent graphic in the famous IPCC Third Assessment Report was the iconic hockey stick graph, which is the foundation of the claims that 1998 was the "warmest year" and the 1990s the "warmest decade" of the millennium, claims that were repeated over and over in promotion of the Kyoto protocol in Canada and doubtless elsewhere. I thought that both the claims and the graph were highly promotional (and this from someone with extensive experience in mining promotions) and began investigating the matter on a casual basis without any expectation that anyone would be interested in my findings. My interest and commitment to the topic would not then have risen much above the level of undergraduate browsing.
My browsing did go so far as to try to identify the underlying data and, for some reason, I contacted the author of the hockey stick study, Michael Mann, when I was unable to locate the data. I initially became engaged in the matter in a more serious way, when Mann said that he had "forgotten" where the data was and one of his associates, to whom Mann turned over the inquiry, said that it was not in any one place, but that he would get it together for me. I drew that conclusion that no one had ever checked Mann’s work and thought that this would be an interesting project, rather like doing a large crossword puzzle. At the time, like any undergraduate reading this, I had never written an academic article. I certainly had no plans to become engaged in academic controversy.
One thing led to another. It turned out that the hockey stick study was a very flawed piece of work and my coauthor, Ross McKitrick, and myself have written several articles criticizing various aspects of the article. Throughout this blog, you will see other issues with the original Mann study and similar studies. Because of the prominence of the hockey stick in IPCC, criticizing the hockey stick has occasioned tremendous blowback. Much of the early life of realclimate was spent trying to preempt our criticisms of the hockey stick. More recently, the "consensus" approach to the matter is that the hockeystick never mattered in the first place.
So where does that leave someone who still wants to know about the impact of increased CO2 on climate – in a ground truth sense, not in a pablumized sense of: here’s what "we know" – not Gavin’s "solid discussion which can be used in another forum", meritorious as that may be.
First, there are many complicated statistical issues. I don’t pretend to be much more than a one-eyed man here. I know enough to be aware of the issues. I’m shocked at the statistical ineptness of people purporting to be climate scientists. The statistical ineptness is quite weird, because, in some climate areas, you see very sophisticated math being applied to deal with complicated physics. But for some reason, this doesn’t seem to be the case with their statistics. As a rule of thumb, for any undergraduates: don’t assume that any of these guys have a clue about statistical significance. There’s been some recent threads at realclimate that illustrate this in spades. For any undergraduates with a strong interest in statistics: there’s a real gold mine of topics in climate science and I would urge you to take an interest in it.
Second, any consideration of climate policy matters will quickly bring you into contact with general circulation models (GCMs). In terms of academic productivity, I have much unfinished business with multiproxy studies, which as a matter of thoroughness, I wish to complete. However, since the promoters of these studies now say that they "don’t matter", some of the edge is being taken off the enterprise. Also by running this blog, one is brought into contact with readers with more general interests than multiproxy studies, as interesting as I may find them.
There are some disquieting points about GCMs. I won’t do anything more than allude to them for now. I’ve posted up about Robert Kaufmann’s finding that, for the purpose of modeling global temperature, GCMs do not out-perform simple linear models using the same forcing factors, and, in fact, under-perform them. Kaufmann posted this at realclimate. Gavin’s realclimate answer was that the issue of global temperature was a "done deal", that the GCMs had "moved on" to regional issues. He requested that Kaufmann continue any discussion of this pretty interesting issue off-line, undoubtedly contributing to the disquiet that the undergraduates feel about them side-stepping issues. Gavin seemed to suggest that those benighted people who were interested in global temperature, rather than "moving on", should look at EBMs (energy balance models).
In any event, for any undergraduates who’s come here, be warned that my posts are pretty uneven – this is just me, not an entire Hockey Team. However, you can be assured that, unlike realclimate, the objective is not to provide you with materials "which can be used by others in different forums", but to inquire about the issues. I hope that you conclude that: when we play hockey, we go into the corners and don’t just dipsy-doodle at center ice; when we play baseball, we step up to the plate and don’t just take practice swings. (Did I mention squash?)
Here is a more recent epistle written in late September:
What is Climate Audit and what is its purpose?
Climate Audit is my personal blog started at the end of January 2005. It has developed rather a life of its own — we get about 10,000 hits a day. In one and half years of operation, I’ve made over 800 head posts and readers have made over 30,000 comments.Initially, I started it purely as a defensive mechanism. Together with Ross McKitrick, I had written a couple of critical articles on the MBH hockey stick for academic journals. While we were waiting for publication of an article in GRL, in Dec 2004, Mann and associates launched a new blog (realclimate) which spent much of its early history on various pre-emptive attacks on us (Yet Another False Claim by McIntyre and McKitrick; Dummies’ Guide to the Hockey Stick, etc.) and attacked us in correspondence to journals.
It became very clear that our articles would never get any sort of hearing unless we responded to these attacks and so I started climateaudit at the end of January 2005, more or less coinciding with publication of articles in GRL and Energy and Environment.
The scope of Climate Audit is a bit different than realclimate or the Pielke sites. While there is a lot of discussion of the MBH hockey stick, I often use the blog as a form of diary on topics that I’m reading about or working on. Thus there are many notes on the various reconstructions published by the “Hockey Team” — Mann, Bradley, Jones, Briffa etc.- and on many individual proxies and even technical statistical issues such as spurious regression. Some of these notes may find their way into a publication some day. For me, since I’m working on my own, the blog functions as a type of on-line seminar.
The readers have a wide range of competence, but include some pretty talented statistics post-docs. Several of the leading authorities in the area have occasionally posted at the blog (Zorita, BàÆà⻲ger, Curry).
The controversy has become very prominent and been widely covered. There are many posts about topics in the news, including ones in which I’ve been involved directly or indirectly — such as the National Academy of Sciences panel on Surface Temperature Reconstructions, hearings by the House Energy and Commerce Committee, to pick two of the most prominent.
The blog has become pretty sprawling. There are many different topics covered, but they developed over the past year and a half not in any particular order. I suggest the following tools: use the Categories on the right frame; that will organize most of the posts. For papers, go to the page on the right frame multiproxy pdf’s — that has the relevant academic literature and presentations. The PPTs provide a type of overview.As to the hockey stick, I’m convinced that no confidence whatever can be attached to any of the reconstructions — which, in my opinion, have been wildly oversold. Some evidence supports the idea that the modern warm period is warmer than the MWP, but other evidence points the other way. I don’t have strong opinions on the matter, but I’m somewhat inclined to the view that the MWP was a little warmer than the late 20th century.
I don’t have any personal views on the hurricane debate. There’s been discussion of this topic on Climate Audit, but nearly all by others. It seems quite plausible to me that warmer SSTs would cause more hurricanes.
I think that there are some real issues with the representation of solar forcing in models, but am not in a position to comment in detail on the matter. I intend to direct some discussion at the blog towards this topic over the next few months.
I might add that I’m a believer in niche markets. This particular blog is not for everyone. It might not be suitable for climate scientists. I don’t know what its “mission” is or even if it has a “mission”. Maybe it’s a bit of a consumers’ guide to paleoclimate science manufacturers.
Climate Audit 
35 Comments
I would add the role of CA in intiating the NAS panel on confidence in reconsructions in the last 2000 years, and in the Wegman report, and might well be mentioned. The start of policy bodies like the House Energy Committee doing their own due diligence in climate science marks a major shift I believe. The expectation of climate scientists, peer review or whatever changing pretty futile, and is secondary to the start of ‘consumers’ of science, policy professionals, conducting their own completely independent inquiry into questions with huge financial impacts.
One interesting thing that has happened to me is a growing trickle of invitations to give a lecture to undergraduates: 3 invites at 2 universities so far this fall. The most recent inviter teaches an upper year survey class on global environmental change, and contacted me because he wanted to prompt his students to be more critical of what they were reading on climate change. Though the prof himself is generally on-side with global warming he had found that on this topic the normally skeptical undergraduate mind was fossilizing, and he asked me to present some challenges to their understanding. It was an enjoyable lecture, some good interactions and no rotten tomatoes.
Last year another prof in a related science area contacted me because he was doing a lecture on the problem of ‘jumping to conclusions’ in science and he wanted some help shaking up the students’ assumptions. I have no idea of his views on global warming, but he was fascinated by the hockey stick episode and he wanted students to take away the message of maintaining a critical attitude.
I have another one coming up in November. I am encouraged that the profs involved are willing to have their students listen to some arguments on the ‘other side’, even though, like I say, none of them are known as skeptics.
Mr. McIntyre, your blog is an exceptional example of how the “new media” can stir up an otherwise complacent and entropic field of study. Often outsiders are needed to point out deficiencies in methods, and faulty assumptions. You are providing an excellent service to the climate community and to the taxpayers of all the countries who are burdened with politically motivated and expensively trendy climate policies.
Now let’s hear something from those students!
Re #2
1. Congratulations on the invitations. Make no mistake, folks: the [penny?] stock of M&M and CA is rising.
2. These professors understand what I’ve come to realize – that hanging out with “skeptics” is good for the mind. You don’t have to swallow what people say here; you’re free to challenge.
3. Students and the public need to understand that if you get all your news from one source – be it right-wing, left-wing, Fox, or RC – you’re not getting a diversity of opinion. We need to become discerning news consumers. And blogs are an ideal tool for discernment.
Re #4
… hopefully in the topical threads.
I wish this technology was implemented back when I was an undergrad. I might have actually decided differently about my career path. Who know, its never too late. I may yet “return home” to Earth Sciences.
#5 — “hanging out with “skeptics” is good for the mind”
In my own practice of science, I’ve found the best way to appraise a topic is to read the contradicting arguments. One discovers very quickly whether the positive idea is sound, or not, by whether the countervailing points have any strength. Science moves, after all, by conjecture and refutation, and not by conjecture and confirmation. And especially not by uncritical confirmation.
In the case of climate science, the work displayed in Steve M.’s blog has shown that proxy paleoclimate reconstructions can yield no high precision temperature-time plots. The fact that climate scientists continue to produce such plots demonstrates a serious failing of scientific perspective. And until someone constructs an actual error metric for GCMs, the theoretical base for valid climate projections remains nil.
Isn’t that just a theory? 🙂
#9 — The description seems to capture the history of science, and so there is likely some empirical support. 🙂
“just a theory” — You put your finger on it, that phrase may be the one most abused by modern anti-science luddites.
My, how things have changed. I recall emailing you when a promised update to the original pre blog website was a couple of months late. Now if I go away for a couple of days I can hardly catch up.
I’d been following Ross M’s website fairly regularly but didn’t find out right away about Steve’s blog. I think I’ve been here every day since except when I’ve been out of town. Perhap’s I’ll go check when I made my first post, which I was wondering about earlier today.
I have really enjoyed this blog. It took me a few months of checking in and reading it to get the gist of M&Ms critique of the Hockey Stick. David Stockwell’s site does a good job of explaining it. I had to laugh when I realized a key aspect fo the Hockey Stick model is the presumption that the models can select which trees are good thermometers for a 1000 year period, based on how they respond to the last 100 or so years of instrumental data.
Re #13 And not instrumental data in their own close surroundings, by the way, but the UHI-contaminated “global average” record. The immediate surroundings of these trees is “rural” and therefore not behaving according to the “standard” global average temperatures of Jones or GISS. Somehow, these particular trees grow faster in response to temperature changes recorded far away while not correlating at all well with temperature changes in their immediate environs.
#2
Congratulations on the speaking engagements Ross. I think, if nothing else, it is heartening that at least in some cases, the professors are challenging their students to think critically about the information that is being presented. Too often, we hear about the flip side of professors who won’t accept thinking that differs from their own. Obviously, one major purpose of education is to teach the subject based on the evidence available. However, it is just as important that students are taught to think, deconstruct, and perhaps most imporantly, challenge the assumptions presented. You may discover that you can’t refute the lesson, in which case you then accept its premise. That seems to me one of, if not the primary goal in science. When refutation fails, hypothesis is accepted as theory. However, even theory should be challenged, as time, tools, and technology expands our capabilities to search for answers.
#13,
I too enjoy this blog. Clearly, the math is over my head, but I certainly get the gist of what’s being said.
Personally, I’d like to thank Steve for starting this blog, and giving me a place where I can try to expand my horizons a bit. I enjoy reading the contributions of (nearly) everyone here. Frankly, I have no idea what the position of most people who post here in AGW here is, and I don’t think for me it really matters. What does matter is getting the science (and math) right. Bad science and bad math should be intolerable. Policy based on bad science/math is usually unbearable.
#13, #14: RIGHT ON, BROTHERS. The tree ring thingie is a fiasco. Except for some real special areas, tree rings tell a precipitation story.
The math is above my head most of the time, but still reachable if you’re prepared to stretch yourself. Steve is nice enough to answer questions to help fill in the gaps.
In my experience, there is no substitute for reading around the subjects raised for oneself – when this blog began, I didn’t have much comprehension as to what significance Steve’s arguments on verification statistics actually had, but I’m substantially less ignorant now.
This advice can apply to everyone, that science and critical thinking is a work that everyone can engage in, not just those who have the sign on the door. In fact, its imperative that more than just professional scientists engage in science.
There is one missing piece about multiproxy studies and GCMs that should be made: multiproxy reconstructions are frequently used to “train” GCMs in the range of various non-physical parameters used in climate models.
So if a multiproxy reconstruction has negligable climatic variation for most of its period and then a sudden shift upwards that is correlated to a single forcing like carbon dioxide, then presto! carbon dioxide becomes the key parameter.
I think the most important result of this effort(blog etc) is:
a) the exposure of the lack of sharing (of data and methods) and replication of at least some of the foundation of AGW theory by scientists who expect policy with trillion dollar price tags to be implemented based on these (unaudited) theories. Unbelievable!
b) peer review is not sufficient where such expensive policy decision are to be made
Re #19
And don’t forget c) that there are additional, more recent multiproxy data that exist, but are being held back for some undisclosed reason.
I hang out here all the time, mostly in the background (lurker). What I appreciate the most are the many paleoclimate posts, from which I’ve learned a great deal. The comments are entertaining and irresistible. Some of them are really clever.
In my office we spend a LOT of time doing quality control (QC) on data. What Steve has become is a really good QC person for paleoclimate studies, which really need someone like him. Willis and Bender and other posters have contributed in this way also.
Paleoclimate is a growing and hugely important scientific field, but it’s still young, unsophisticated, and badly in need rigorous quality standards. Steve and Ross have advanced the field significantly through their insistence on reproducibility and high quality. We all owe you a big debt of thanks!
#16 I would think tree growth might be a function of several things, say precipitation, sunlight, temperature, CO2. At differnet values of these variables, a tree might have no response to a change in one and big response to change in another.
#22, Not to mention, infestation from insect/mold/spore/fungii, encroachment of other species (both plant and animal), near urbanization of encroaching civilization on rural areas, etc. There are so many factors involved, how anyone can decide that there is a high level of confidence in tree-rings as a temperature proxy is beyond me.
sure there are confounding factors. But there are also confounding factors with consumers and marketing science exists. There are confounding factors with human lives and accidents and insurance and actuarial science exists. There are confounding factors in manufacturing processes and people still work to make them better. Etc. Etc. Etc.
#24,
Yes, there are various factors that affect statistics in a variety of businesses. However, the question is whether those are primary, secondary, tertiary, or other, and so their impact may be important or may be negligble. It’s up to the people who develop the statistical secenarios to determine the relative importance of those factors, and then look at how they may or may not impact the results. However, the statistics used in insurance, marketing, etc., is also based on the historical records available, plus various demographic data that is injected into the formulas. In the case of tree-ring proxies, in many cases, there is no historical record, other than other proxies or pseduo-proxies. That would to me indicate a much lower confidence level in ascribing any one particular factor as the primary driver in the growth of tree rings.
Reading this particular thread and all the back-patting there is no shortage of amusement because of course, you’re all too late because “it didn’t matter all that much anyway.” Which would have been interesting to hear said only a few of years ago when many were busy trying to ensure that everyone was on board with the decided course of action. I can only imagine what a late 1990’s international conference would have been like with a certain piece of evidence removed for being unsupportable. Would people have said back then that it “didn’t matter” ??
When the main remaining tactic of your “opponents” is to deny the existence of an argument (oh, it never mattered anyway), doesn’t that mean that at least in that niche, that the “debate is over” (so-to-speak of course)??
“JUST BECAUSE WE’RE BEREAVED DOESN’T MAKE US SAPS!!!” — Walter Sobchak, The Big Lebowski
24: Yeah, yeah, yeah. BUT, we don’t presume to estimate the effects of these factors 1000 years ago. TCO, you are groping.
TCO, you misinterpret the subject of the concern when you say in #24:
The problem is not with the confounding factors. As you point out, these exist everywhere, including all the sciences.
The problem is the extreme reluctance of the AGW crowd to put realistic uncertainty estimates on their historical reconstructions and their forecasts based on these confounding factors. Here’s an example.
James Hansen et al., in the abstract of their “smoking gun” paper “Earth’s Energy Imbalance: Confirmation and Implications” that was supposed to “prove” AGW, say the following:
Now, your average journalist reading this, or even your average scientist, would assume that the “±0.15” is some kind of an uncertainty estimate regarding the model results … but it is no such thing. It is the difference they obtained from five model runs, which means nothing. Nowhere in their paper do they offer any kind of error bars or confidence interval for their model results,
This, of course, may be related to the fact that their model results compared to actual data have a 95% confidence interval of ±1.0 W/m2 … which is not only more than nine times their quoted error, but is larger than their result as well.
And when errors or confidence intervals are quoted, most of the time they are only expected statistical errors in a reconstruction, or differences in model runs as in the example above, rather than an estimate of the actual error (results vs. reality). This practice, intentional or not, isvery deceptive, particularly to the media.
That lack of error bars or confidence intervals in the overwhelming majority of AGW studies is the problem, not the confounding factors.
w.
It has been half a century since I took statistics. Perhaps you can help me with a simple question. If a data series has measured values of 1,2,3,4,& 5 and each measured point has an accuracy, variance or error of +-5, what can I derive from this series? Is there a trend?
thanks.
I just wanted to make the distinction that the students you are referring to from GaTech are not undergraduates. Ill revisit what Dr. Curry said a few times in the Report Card:
I, as well, have already received my BS in meteorology.
Im not sure that this necessarily changes the way you could read this post, but I think its important to understand the backgrounds of the “reviewers” at hand. There is a solid background in science/statistics there.
Re: #30
Angela F. in response to:
Says in part:
I agree that there may be “maturity” issues here but that those issues have a bearing on the final conclusions of either group of students I have my doubts, since I didn’t see any confirmation in Blumenfeld’s student reactions to the RC and CA debate of preferences for what goes on here, only that they wanted RC to get more down and dirty with the CA heretics on AGW and the HS. Being undergraduates they might express their feelings with a more live or die by the home team attitude that requires them to get more down on the home team when it fails them. In the end, I view both groups as probably not having much of a liking for what goes on at CA.
My problem is that without a dialogue with these students on specific issues it is difficult to determine the reasons behind the dislike of CA and it would also make me hesitate, no make that refuse, to pass judgment on the motives or thinking of the students.
#24,
There are confounding factors in manufacturing processes and people still work to make them better.
I must have missed all the posters here that are firmly in the “we’re against working to make science better” camp.
There are two problems with your comment.
The first is that Mann and others seem to be the ones resisting working to make science better not the people here.
Second, and more important, prior to any meaningful knowledge of the scope of these confounding factors the world is being asked to turn itself inside out economically based on the rudimentary and almost undoubtedly highly inaccurate climate projections produced so far.
#40 Angela,
The title of this post is “To Browsing Undergraduates”. I’d be a bit pissed too at first glance. This was a screw-up on Steve M’s part, no doubt. If he reads this I’m sure he will realize the error.
Dr Curry has taken worse here, especially early on, and IMO has benefitted by it. I appreciate that aspect about her.
FYI, to the rest of you, GT is ranked about 3rd or 4th in the US for engineering schools. It is no small feat to get accepted into a grad program there.
Good luck on your studies, Angela.
Don’t so sensative, you little flowers, you. The earlier thread was for UGs. This is a reposting of that thread. You can still learn from it, even though it was first composed for dorm-living UGs.
Re: Ken
I would agree with that.
I would definitely agree with that. We were just discussing yesterday how much CA or any other blogging atmosphere helps you “fine tune” your critical thinking skills, as well as the ability to defend research.
Thanks, Joel. 🙂