Saturday, December 05, 2009

There are two sides along De-Nial

Local reaction to ClimateGate reflects exactly what the scandal itself illustrates, i.e., that the question of climate change has become almost hopelessly politicized. Some of my friends on the right are making claims that are a step or three too far. ClimateGate is indeed a scandal, but it doesn't disprove claims of AGW or demonstrate that they are fraudulent. Rather it reflects something that skeptics have been saying for a long time. The science is not "in." The business of determining the impact of human activity on the climate and predicting what will happen in the future is frightfully difficult and climate scientists have made relatively little progress. There is, at best, awfully thin support for the notion that AGW is an existential threat that requires hobbling the world economy in the ways avocated by President Obama and former Vice President Gore (who has become the world's first global warming profiteer.)

On the left, bloggers like Illusory Tenant are redefining the term "denialist." There is, they say, nothing to see here and "
wing nuts" who say other things are so stupid - not at all like us smart people.

No, actually using a "trick" to "hide the decline" is a big deal. The problem is not that no one could figure out what they were doing (although it apparently took Michael Mann a while to figure it out). Steven McIntyre was certainly aware of it and unsuccessfully tried to persuade the IPCC not to do it.

The problem is that it reflects a willingness to assume away inconvenient facts and then to construct graphs for broader public consumption that "hide" those facts and that are, taken by themselves, fundamentally dishonest. Combining proxy data with actual measurements is a "trick" in the bad sense of the word. It acts as if apples and oranges are the same thing.

More fundamentally, it reflects a bias toward a particular conclusion. Apparently tree ring density was a good proxy for recorded temperatures for the first 60 to 80 years for which actual temperature measurements are available. After at least 1960, it is not. When attempting to make a claim about temperature trends over centuries, this is a problem. How can we know whether the period before 1880 is like the period from 1880 to 1960 or like that since 1960? How can we know that the relationship between tree rings and temperature is constant over those hundreds of years?

If we don't know that, then how can we know that the observed rise in temperatures over the past one hundred years or so is not simply part of normal climate fluctuations? Good science does not respond to the problem by calling it a "travesty" and acting as if it does not exist. Good science, as McIntryre points out, requires that you acknowledge the problem - including in documents that you want policymakers to rely on like the IPCC reports - and deal with it. That insiders knew about the issue does not justify deceptive representations of the data to the larger community.

The discarded data is a problem. It's not that there aren't other data sets and the raw data used by CRU might be reconstructed, but it was an important, widely used and much adjusted data sets. Now the adjustments can not be readily examined. Given what the e-mails tell us about problems with the data set and the raging partisanship revealed by the e-mails, this is a rather large problem.

One of the best summaries of this has been written by Steven Hayward in the Weekly Standard. Here's his conclusion:

Climate change is a genuine phenomenon, and there is a nontrivial risk of major consequences in the future. Yet the hysteria of the global warming campaigners and their monomaniacal advocacy of absurdly expensive curbs on fossil fuel use have led to a political dead end that will become more apparent with the imminent collapse of the Kyoto-Copenhagen process. I have long expected that 20 or so years from now we will look back on the turn-of-the-millennium climate hysteria in the same way we look back now on the population bomb hysteria of the late 1960s and early 1970s--as a phenomenon whose magnitude and effects were vastly overestimated, and whose proposed solutions were wrongheaded and often genuinely evil (such as the forced sterilizations of thousands of Indian men in the 1970s, much of it funded by the Ford Foundation). Today the climate campaigners want to forcibly sterilize the world's energy supply, and until recently they looked to be within an ace of doing so. But even before Climategate, the campaign was beginning to resemble a Broadway musical that had run too long, with sagging box office and declining enthusiasm from a dwindling audience. Someone needs to break the bad news to the players that it's closing time for the climate horror show.

18 comments:

illusory tenant said...

On the left, bloggers like Illusory Tenant are redefining the term "denialist."

You just made this up. I accept it as a term of art, and neither defined nor redefined it. I do understand it's not complimentary. By the same token, "alarmists," "tree huggers," and even "ManBearPig" appear equally often in the literature.

There is, they say, nothing to see here and "wing nuts" who say other things are so stupid - not at all like us smart people.

No, I never said any of those things (except "wing nuts," which is apt: see the claims of Limbaugh -- "wing nut" central -- et al). Obviously there is something to see here and you should not infer so much from style.

And nearly everything I have said about this topic has been in reaction to its "reporting" in the local paper.

By its editorial board member emissary, the paper has been repeatedly pushing baseless right-wing memes, and it's long since evident that their author either has no understanding of the underlying facts or else is deliberately playing stupid.

Since you don't provide links, here is the blog you mentioned.

Anybody who could publish this:

"One telling message from the unit's head is about how to 'hide the decline' in observed temperatures, as global warming seems to have halted about a decade ago"

has no clue what he's even talking about.

Reconstructing global climate over millenia is a massive and complex undertaking. If the most damning thing you guys can come up with is a substitution of instrument for proxy measurements (by the way, the decline at issue -- which is a decline in the proxies' temperature sensitivity, not a decline in temperature -- is noted in a carefully identified selection of trees, the "negative responders") on the front of a convention brochure to account for a well known and widely discussed disagreement between two separate sets of data, then you don't have much at all.

Most significantly, however, is that adjusting for the statistical bias introduced into long-term climate reconstructions by the divergence problem could show historical temperatures to have been cooler than estimated, and therefore underepresenting the currently acceptable expression of "alarmism."

Clutch said...

No, actually using a "trick" to "hide the decline" is a big deal.

I was recently at a PhD defense in experimental psychology in which the candidate talked about the "trick" he used to get the statistical results to emerge more clearly from the data. The external reader suggested that the phenomena might have been even clearer if he'd used a different "trick".

OMG psychometrics fraud! If only I'd known it was a big deal!

I guess my mistake was in knowing what they were talking about. One should really be armed with ignorance if one wants to be able to recognize a big deal when one sees it.

jp said...

Clutch;

Are you suggesting the PHD “trick” used openly is the same as the hidden emails?
Excuse me I am going to wash now.

Clutch said...

Oh, the PhD stats "trick" remark was every bit as "hidden".

It took place in quite a small room, with the door closed, with only about 6 people listening. (I myself was one of 4 cc'd on it, as one might say.)

OMG we all have to wash now! Soiled by hiddenness and "tricks"!!

illusory tenant said...

I see it took your friend Patrick McIlheran two weeks of continual misrepresentation to find out -- from some conservative blog -- that "decline" did not mean "decline in temperatures" which he now casually dismisses as merely "a subtlety" that he'd somehow overlooked.

So who's the one in denial again?

Rick Esenberg said...

The problem, Tom, is that the claim for the measured warming being unusual is inextricably tied to proxy data because there were no measurements. If the sensitivity of the the proxy data is called into question, then the claim of historically unique warming is called into question. This, it seems, to me cannot be dismissed as "not much" although I do think that the e-mails emphasize - rather than reveal - it. We knew about the problem all along even as the alarmists refused to acknowledge it.

Now, if I wanted to, I could argue that your apparent failure to acknowledge the obvious (your own form of denial which I did not make up) suggests all sorts of bad things - that you are or are playing stupid or have no credibility or are not addressing the issue in good faith. Perhaps that would be a matter of style.

As far as Clutch's comment, the use of the word "trick" is embarrassing, but the problem is not the use of the term nor do we need the term to alert us to the problem - which is combining apples and oranges to hide the inability to demonstrate a historically unique warming trend.

Grant said...

If the sensitivity of the the proxy data is called into question

This use of definite articles achieves precisely the bait-and-switch effect you're decrying. You did the same thing in comments to a previous post: "RealClimate talks about deniers and disinformation and hangs on to certain claims (i.e., the hockey stick) which most other scientists seem to have abandoned."

It would be helpful if you specified which data and which hockey stick to which you're referring.

illusory tenant said...

If the sensitivity of the proxy data is called into question, then the claim of historically unique warming is called into question.

We know that, and have known for years. And yes, not only did I acknowledge it, I addressed it:

"Adjusting for the statistical bias introduced into long-term climate reconstructions by the divergence problem could show historical temperatures to have been cooler than estimated, and therefore underrepresenting the currently acceptable expression of 'alarmism.'"

Just in case you missed it.

And we know that a portion of Briffa et al's data was set aside in favor of a contemporary series of instrument measurements for the cover of a convention brochure, a brochure which also explained -- as opposed to "hid" -- that the graph was built from a composite of instrument and proxy measurements.

And we know that Briffa et al's and some others' data show that during a certain period of time, certain northern hemisphere trees demonstrate a reduced sensitivity to changes in temperature and thereby growth indicators that do not positively correlate with instrument temperature measurements for that period of time.

And we know that there could be a number of explanations for this, and we know that these explanations will affect the longer-term reconstructions which necessarily use exclusively proxy measurements. (The latter is blindingly obvious; I fail to see how it's even possible to deny.)

And we know that there is a good deal of other research that shows no divergence, or considerably less pronounced divergence, among other populations of trees not just at lower latitudes, but also among those comparable with Briffa et al's Alaskan samples, that is, those within the highest northern hemisphere latitudes.

And there is other research that shows that the divergence of greatest concern (that identified by Briffa et al in 1998) is unique to recent history, which in turn implies that the fluctuation in sensitivity to temperature among certain northern hemisphere trees is itself of anthropogenic cause.

And we know that there have been a number of temperature reconstructions since 1998 by other researchers using entirely separate populations of trees that essentially confirm the earlier reconstructions.

And we know very well that far from "debunking" AGW as a "hoax," the most compelling explanations for the divergence problem may show a wider disparity among temperature anomalies over time than even the convention brochure cover did.

Finally, we know that all of the temperature reconstructions at issue together comprise but one line of evidence among many presented for the so-called "alarmist" propositions. Remove this one line of evidence from the array, and the case is no less compelling.

We know all this. Or, to put it another way, I have never denied it, which is your apparent accusation. Now if you can't point to me denying any of these things (those things that I just affirmed above) then your case for my being a "denialist" is without any merit.

Yet you simply assert it once more.

So I'll ask again: Is this really all you guys have? "Hide the decline"?

Which it took Patrick McIlheran two weeks to only just now begin to figure out and he still hasn't adequately contemplated before rushing headlong onto the pages of the Milwaukee Journal-Sentinel to announce even more ill-informed declarations of science?

You gotta be kidding me.

George Mitchell said...

The extent of warming, and man's contribution thereto, is not settled science. This was clear prior to Climategate. The e-mails serve merely to confirm that. There are political activists, including some Milwaukee bloggers, who bring a smidgen of scientific knowledge to the table and claim that in fact the science is settled. They would not last 5 minutes in a discussion with reputable scientists who have raised valid concerns about the science.

George Mitchell said...

Research evidence regarding school choice provides some interesting context. There is general agreement among the most credible scholars that low-income students, especially Blacks, gain from choice. About twelve studies in highly regarded scholarly journals support this finding. The data and methods used to reach these conclusions are widely available and have been presented and discussed at countless academic forums. Aside from some laughable spin from some partisans, the debate as to whether Blacks gain is settled. The magnitude of those gains is subject to varying interpretations. Whether those gains are sustained requires more research. The credible scholars in this field do not find themselves in the kind of dispute surrounding some of the Climategate researchers. The leaked e-mails do not cast a positive light on those researchers.

Anonymous said...

I think the better question is; why are progressives trying to make every single person responsible for controlling the weather?

Do they really think that would have stopped the glaciers melting from North America?

illusory tenant said...

George, who said 'the science is settled'? It wasn't ManBearPig by any chance, was it? Also, how could social science possibly be settled when it's not even possible for physical science to be settled?

George Mitchell said...

I have no clue as to what IT is referring or pursuing. Is he/she suggesting that research on educational programs is inherently inconclusive?

As for school choice, the most demanding research criteria involves randomized assignment of students to choice and non-choice environments, i.e., the "gold standard" used for drug trials. Research using that criterion repeatedly finds that Blacks gain from having choice. I would be glad to provide links, but my guess is that IT could not care less.

Terrence Berres said...

"who said 'the science is settled'?"

Headline writers, perhaps: Copenhagen: The Science Is Settled; The Policy And Politics Aren't, by Marc Ambinder, Politics blog, The Atlantic, Dec 7 2009, 1:13 pm

Jay Bullock said...

Let's be fair about the voucher research, George:
The primary finding in all of these comparisons is that there is no overall statistically significant difference between MPCP (voucher) and MPS student achievement growth in either math or reading one year after they were carefully matched to each other.

George Mitchell said...

Jay,

Agree..Let's be "fair."


Random Assignment School Choice Studies

1) Study: The Effect of School Choice: An Evaluation of the Charlotte Children’s Scholarship Fund Program.

Author: Jay Greene, University of Texas.

Published: Education Matters (2001)

Location: Charlotte

Findings: Significant academic gains in math and reading.

2) Study: School Choice as a Latent Variable: Estimating the "Complier Average Causal Effect" of Vouchers in Charlotte.

Author: Joshua Cowen, University of Wisconsin.

Published: The Policies Studies Journal (2008)

Location: Charlotte

Findings: Significant academic gains in math and reading.

3) Study: School Vouchers and Academic Performance: Results from Three Randomized Field Trials - Dayton

Authors: William Howell, University of Chicago
Patrick Wolf, Georgetown University
David Campbell, Notre Dame University
Paul Peterson, Harvard University

Published: Journal of Policy Analysis and Management (2002)

Location: Dayton

Findings: Significant academic gains for African-American students.

4) Study: School Vouchers and Academic Performance: Results from Three Randomized Field Trials - Washington D.C.

Authors: William Howell, University of Chicago
Patrick Wolf, Georgetown University
David Campbell, Notre Dame University
Paul Peterson, Harvard University

Published: Journal of Policy Analysis and Management (2002)

Location: Washington D.C.

Findings: Significant year 2 academic gains in math and reading with higher gains for African-American sub-group in year 2.

5) Study: School Vouchers and Academic Performance: Results from Three Randomized Field Trials – New York

Authors: William Howell, University of Chicago
Patrick Wolf, Georgetown University
David Campbell, Notre Dame University
Paul Peterson, Harvard University

Published: Journal of Policy Analysis and Management (2002)

Location: New York

Findings: Significant academic gains for African-American students.

6) Study: Effectiveness of School Choice: The Milwaukee Experiment

Authors: Jay Greene, University of Texas
Paul Peterson, Harvard University
Jiangtao Du, Harvard University

Published: Education and Urban Society (1999)

Location: Milwaukee

Findings: Significant academic gains in math and reading. Gains were highest after several years.

7) Study: Private School Vouchers and Student Achievement: An Evaluation of the Milwaukee Parental Choice Program

Authors: Cecilia Rouse, Princeton University

Published: The Quarterly Journal of Economics (1998)

Location: Milwaukee

Findings: Significant academic gains in math.

8) Study: Principal Stratification Approach to Broken Randomized Experiments: A Case Study of School Choice Vouchers in New York City

Authors: John Barnard, The Cleveland Clinic Foundation
Constantine E. Frangakis, Johns Hopkins University
Jennifer L. Hill, Columbia University
Donald B. Rubin, Harvard University

Published: Journal of the American Statistical Association (2003)

Location: New York

Findings: Significant academic gains in math for African-Americans and students from low-performing schools.

9) Study: Another Look at the New York Voucher Experiment

Authors: Alan Krueger, Princeton University
Pei Zhu, Princeton University

Published: American Behavioral Scientist (2004)

Location: New York

Findings: No significant academic gains.

10) Study: The Evaluation of the DC Opportunity Scholarship Program: Impacts After Three Years

Authors: Patrick Wolf, University of Arkansas
Babette Gutmann, Westat
Michael Puma, Chesapeake Research Associates
Brian Kisida, University of Arkansas
Lou Rizzo, Westat
Nada Eissa, Georgetown University
Marsha Silverberg, Institute of Education Sciences

Published: U.S. Department of Education (2009)

Location: Washington D.C.

Findings: Significant academic gains in reading.

Anonymous said...

Disproving global warming is almost as difficult as disproving the existence of god. Maybe you should blog about that.

illusory tenant said...

As I was saying.