I much appreciate all the responses to my earlier question about data sharing in the climate change debate. The quantity and the detail of the responses was dauntingly high, and I’ve just managed to get through them all. I thought therefore that I’d post a bit more about the subject.
1. It looks like some of the data sources — apparently government-run meteorological services — did indeed demand (whether contractually or as an informal understanding) that the data they gave to CRU not be shared further. Apparently this is because the services sell the data commercially. But to the extent that CRU’s continuing refusal to share the data stemmed from those agencies’ refusal, this just means the problem is broader than I had thought.
The countries in which these meteorological services are located will have to, in the aggregate, spend trillions of dollars under various climate-change-fighting proposals. If I’m right that data sharing is an important part of making science accurate, those countries have much to gain from such sharing.
Even if the global warming claims are right in general, it’s nearly certain that they’re not right in all particulars; that’s not a condemnation of global warming theory, but just a reflection of what happens with nearly all emerging scientific theories. The particulars might well matter in deciding on the best way to combat global warming. An error in the models might yield errors in the recommendations; public access is more likely to uncover those errors.
Of course, if the global warming claims are wrong in important respects, the countries involved have even more to gain from their being proven wrong. And if they are largely right, the countries have much to gain from public confidence in the claims’ correctness, confidence that I think should be reinforced by data sharing and undermined by data concealment.
So if indeed the meteorological services are refusing to have data shared because they sell the data, that is astoundingly penny-wise and pound-foolish. Perhaps CRU might justify its actions by saying that the services conveyed the data under a confidentiality understanding. For that we’d need to know more details about the understandings, and also the details of English Freedom of Information law that may or may not preempt such understandings. But where’s the adequate justification for the underlying continuing insistence by government-run meteorological services that their data remain confidential?
2. Some commenters have pointed out that in many research projects, the researchers try to keep their laboriously gathered datasets private for some time in order to milk the data for more papers, without competition by others. I’m not sure whether this is on balance beneficial to science, or to what degree it is tolerated in various scientific communities.
But even if it makes sense for some researchers to trade off verifiability and credibility for extra lead time to produce novel work — and for some disciplines to trade off the benefits of public access for the benefits of extra incentives to gather data — can that possibly apply here? This isn’t just cutting-edge science in which scientists can say, “OK, if you don’t believe my first article, fine; that’s a tradeoff I’m willing to make in order to get more publications on my resume, and then in several years I’ll release the data and dispel your doubts.”
This is research that people are claiming should form the basis of multi-trillion-dollar regulations and expenditures — that should lead to compulsory restructurings of vast segments of society — and the sooner the better. The need for accuracy is so great here that I don’t think the “I want more lead time to write more papers without competition” argument can apply (nor do I think it’s even made by CRU).
3. Some commenters have argued that a great deal of climate data is indeed available from other sources; and that’s good. But as I understand it, the CRU didn’t get its standing by just being duplicative of other data gatherers. The CRU dataset has apparently formed the basis of a great deal of work, and even if it has been seen by many as entirely consistent with the other datasets, it remains important.
And to verify the CRU’s work it’s necessary to have the precise data on which the CRU relied, and not just other data from other sources that has its own limitations and characteristics. Small differences in coverage (temporal and spatial), as well as differences I the methods through which the data was gathered, can potentially make substantial differences in result. And absence of the precise data, with its precise temporal and spatial boundaries, makes it impossible to verify the particular results that CRU reports. If you are so important, and your dataset is so important, it’s important that other scientists who want to check your work have access to your dataset and not just to other datasets.
4. Finally, one commenter asks whether “this work[s] both ways,” and whether “anti-AGW partisans” should share their data as well. Obviously so, and I take it that if someone criticized conventional scientific wisdom (which global warming has, descriptively, become) and then refused to share the underlying data, he’d have zero chance of persuading anyone, and rightly so. The question is how much extra slack should be cut to the researchers whose work forms the conventional scientific wisdom. Again, where many trillions of dollars are at stake, it seems to me the answer should be “not much.”
HarryEagar says:
Concerning 4., Steve McIntyre, the godfather of transparency on climate, shares his data. He posts his code on his blog.
So, yeah, it works that way. Has for a long time.
Quote
December 1, 2009, 5:40 pmspostrel says:
There is also a great deal of question about whether CRU’s appeal to intellectual property restrictions was sincere. Phil Jones, in one of the leaked emails, explicitly referred to “hiding behind” IP claims to avoid giving the Climate Audit crowd access to the data. Furthermore, numerous people have filed FOIA requests to pin down exactly which data are covered by such agreements; the CRU crew have refused to reveal even that much to date.
Quote
December 1, 2009, 5:45 pmDavid Schwartz says:
I guess my counter-argument would be that either the CRU results are robust and appear from the majority of valid data sets with the majority of valid methodologies or they are no robust and rely on specifics of the CRU data set and the specific steps CRU followed. Either way, the CRU data is not needed. The only exception would be if people want to rely on specific aspects of the CRU data that aren’t replicated in other independent data sets.
So in some future, when we have a few provably independent data sets with replicatable results from following a documented methodology, if they confirm the CRU results, that we can’t replicate the exact CRU results won’t matter.
In other words, that we were in a situation where the accuracy of the CRU results mattered was the problem. We should be in a situation where it doesn’t.
Quote
December 1, 2009, 5:47 pmfwb says:
Sharing/not sharing is not the key question. Scientists share or do not share for a number of reasons. The primary issue is that original data was destroyed and the manipulated data was kept. No good scientist destroys original data. The manipulated data can be regenerated from the original info any time.
Quote
December 1, 2009, 6:04 pmsteve says:
http://www.chron.com/commons/readerblogs/atmosphere.html?plckController=Blog&plckScript=blogScript&plckElementId=blogDest&plckBlogPage=BlogViewPost&plckPostId=Blog%3a54e0b21f-aaba-475d-87ab-1df5075ce621Post%3ade6fb89a-bf98-4503-a332-344214aa9a9b
A brief bit from above. Dr. Nielson-Gamon is a Texas State climatologist.
.
I read McIntyre fairly regularly. He does publish his code. Strangely enough, he has not published his emails.
I think you are right on track with number 4. Let us look at all of the data and everyone’s emails. Let the charges of hypocrisy begin.
Steve
Quote
December 1, 2009, 6:15 pmBlue says:
Eugene, I don’t think the CRU was under any obligation to share true raw data. (I emphasize was intentionally.)
Think about the work process from a very abstract level:
1) Raw data;
2) Transformation/cleaning (ALL raw observational data requires this);
3) Useful data set.
What CRU had an obligation to provide was:
1) The specific sources of the raw data;
2) The specific steps used to transform and clean the data–specific enough so that they can be reproduced by other parties;
3) A full list of caveats and limitations on the useful data set that the transformations and cleanings imposed.
This would have allowed third parties to, say, take a sample of these weather stations, pull the raw data, and check the result against CRU’s work. They could also have made different assumptions in the transformation/cleaning process and done a sensitivity analysis of the resulting overall changes in the useful data set.
This approach would also have allowed users to segment off parts of the CRU data set for analysis–for example, only looking at that subset of stations that had continuous records since 1900, had not moved location, and were not in an urbanized area.
The result would have been a series of papers that modified, enchanced, limited, and otherwise led to a much deeper understanding of the underlying dynamics in long term weather trends. None of this was possible, however, given the way that CRU handled its data...and now its too late. The whole thing needs to be recalculated by a neutral party or parties.
Quote
December 1, 2009, 6:24 pmef says:
Since all of this data is “adjusted”, has the general acceptance of the CRU data (compounded with any lack of transparency about those adjustments) resulted in those other data sets adopting “adjustment” criteria that brought it in line with the CRU’s data sets? The possibility for cross project taint in this regard seems extremely high. The AGW needs to go back to step 2 and reassess, independent of the CRU data those assumptions on which the data rests.
Quote
December 1, 2009, 6:27 pmJones' Cell Mate says:
Would it be fair to have McIntyre only turn over e-mails that he destroyed in an effort to avoid lawful disclosure? I’m willing to reach a reasonable compromise on this issue.
As to steve’s initial point, I believe comment #82 in this link addresses it squarely. If steve, or any other commenter, has a rebuttal to that comment– I’d be interested to hear it.
Quote
December 1, 2009, 6:31 pmLester Hunt says:
Eugene: Very sensible and fair, as usual.
Quote
December 1, 2009, 6:43 pmArthur says:
A couple of largely unremarked points which I think have quite reasonably been fueling the skeptics’ concerns:
(a) Most commentators seem to assume that the problems revealed in these emails reveal all the weaknesses in the AGW case that we ought now to be worried about. Yet given the single, very incomplete leak we have so far this is surely unlikely. You don’t have to be much of a conspiracy theorist for instance to think it quite likely that Michael Mann’s correspondence will eventually turn out to contain many more examples.
(b) Given the way this information has emerged, plus the fact that the skeptics have been pushing for much of what is now being revealed for many years, it is surely clear that the entire scientific governance process in this field is broken (precisely as the skeptics claimed). We ought to remember that, absent one well-placed whistleblower, we would still be being successfully kept in the dark by this cabal of conspiring academics.
Quote
December 1, 2009, 6:48 pmzuch says:
Prof. Volokh:
What data? The primary activity of “anti-AGW partisans” seems to be to criticise and nitpick GW studies of others, and not to do any original work themselves (or at least any such work that stands up for long).
Cheers,
Quote
December 1, 2009, 6:54 pmsteve says:
Jones– I had thought that they published all of the emails, not just the ones they attempted to destroy. Am I wrong?
As to commenter 82, he is right in a broad sense. In the ideal world, people would save and turn over everything. They did not for what may be practical reasons, storage, and common practice reasons, as noted by another climatologist above. The point still remains that the data is or should be available. The CRU was not an archival site.
Skeptics claim that they have real scientists on their side. They do. So, rather than spend all of their efforts on discrediting the other side, why not gather data, write their own code, make their own model and publish. IOW, do their own research. Granted I am just a humble (sort of) physician and medicine has its own weakness, but we redo studies all of the time. We criticize freely, but new studies are made. Then we compare to see which drug company financed research is best.
Quote
December 1, 2009, 6:58 pmFourier says:
Keep in mind that this isn’t a case of some simple data set somewhere that someone plots up. There is no global thermometer. People can’t even decide how to average a 24 hour temperature at one spot to represent “the” temperature at that spot. Plus, the measuring sites are moved, corrupted by having parking lots built near them, or inactivated. See SurfaceStations.org for data on the mess that is the US temperature recording system. See <a href=“http://chiefio.wordpress.com/2009/11/09/gistemp-a-human-view/this for E.M. Smith’s useful discussion of the nonsense that goes on trying to build the NASA temperature data. He notes that “For California, where we once had thermometers in the mountain snow and in the far north near Oregon; there are now 4 surviving thermometers near the beach and in the warm south. But GIStemp is sure we can use them as a fine proxy for Mount Shasta with its glaciers and for the snows and ice of Yosemite winters.
CRU constructed global temperature series. It didn’t say how it did it. It supposedly build models from them. And didn’t say how the models worked. All of the work is suspect, right down to the original data.
Quote
December 1, 2009, 7:01 pmJones' Cell Mate says:
Skeptics claim that they have real scientists on their side. They do. So, rather than spend all of their efforts on discrediting the other side, why not gather data, write their own code, make their own model and publish. IOW, do their own research.
As a general matter, the skeptics aren’t advancing the relevant claim. When one provides a theory (AGW) that person/group assumes the burden of supporting that theory. Not the reverse. “Skeptics” have it easy, they have only to find a single failure to reasonably refute the theory.
That’s why science is hard.
Quote
December 1, 2009, 7:11 pmDoctor Gator says:
Minimum standards for open scientific inquiry, loosely stated are:
It must be possible for third party scientists to verify (1) that raw data is valid; (2) that the data values used from a raw data set is not cherry-picked or altered; and (3) that data processing by the third party, applying details of the researcher’s theory and the researcher’s assumptions (aka parameters), produces the same results as the researcher’s processing.
For the above to be possible the researcher must reveal (1) in detail every data value used as input into the process; (2) every detail data value used that represents assumptions; and (3) the precise source and identity and description of segments of the raw data set used. The raw data set must be available to bona fide third party researchers. Data sets used to research public policy issues must be available to the public.
From Steve’s Quote of Dr. Nielson-Gamon:
Dr. Nielson-Gamon’s makes no statement about producing for third party evaluation all data values (input variable and parameters) he uses as process input
Hadley CRU has refused to produce any of the above. As far as I know Hadley has not even precisely identified the raw data sets used (with segments defined) for each climate change profile used in its contributions to IPCC reports or to its regularly published climate assessments. Furthermore Hadley uses raw data sets created internally.
Quote
December 1, 2009, 7:22 pmDavid Schwartz says:
Because most of those skeptics believe that we do not have the ability to produce useful climate prediction results. This is a negative that can’t be proven by trying and failing. All you can do is shoot down purported counter-examples.
Prove that it’s impossible to predict with 98% accuracy in which major cities it will rain precisely two weeks from now.
Quote
December 1, 2009, 7:25 pmsureyoubet says:
It’s the rough equivalent of a lawyer coming to court with a case based on a statute that he claims is proprietary and can’t be disclosed to the judge, witnesses that are only available through summarized declarations, and caselaw that is copyrighted and therefore can’t be shown to the judge, but, trust him, it is summarized nicely in his brief.... (and it turns out some of the witnesses are now dead and the original caselaw has been destroyed.)
How much slack would the judge cut? “Not much.”
Quote
December 1, 2009, 7:28 pmzuch says:
Jones’ Cell Mate:
They did. They published. Not a “theory” but scientific papers (peer-reviewed to boot). Now it’s up to you to show how they got it wrong. It’s not sufficient for you to say, “I don’t trust you; I don’t believe you [and I don’t want to hear what you’re saying, perhaps]. Now show me; convince me; prove it to me....” Such an attitude among the critics may be responsible in part for the animus against them show by the researchers....
Cheers,
Quote
December 1, 2009, 7:42 pmcirby says:
So when all of those scientists around the world criticized cold fusion, they were in the wrong?
Gee — that’s... exactly opposite the way science is supposed to work.
Science isn’t only “prove this,” it’s much more often “find holes in this other scientist’s research.”
You don’t have to repeat someone else’s experiment to disprove the results. More often, you can show an alternate mechanism that explains the flaws in the first experiment, or show how they misinterpreted data to get their result. Some times, you can even prove fraud (often by statistical analysis of raw data — faked numbers aren’t random enough).
But one of the things you always need is the original data. Not “go out and find it yourself,” but “here’s a copy of the database, see if you can find a flaw.” What we’re getting from the CRU is “here’s a copy of our results, no you can’t have the data, no we won’t give you a list of the places we got all of our data from, and no we won’t tell you how we massaged the data to get the result.”
Quote
December 1, 2009, 7:44 pmzuch says:
No. In most cases, the standard practise is for others to reproduce the results themselves (or attempt such). If the two match, now you have even stronger evidence with twice the sample size. If not, they sit down to try and find out why the differences occurred (which may even lead to more interesting observations or refinements of the theory). If someone cherry-picked their data, reproductions by other that do not do this will fail (and this in many cases then lead to the investigations that uncover fraud where such has occurred). In rare cases, where the accounts of experimental procedure seem perhaps to be incomplete or inaccurate and the results are at variance with efforts by others to reproduce the results, outsiders may step in to monitor a replication of the experiment by the original lab (see, e.g., the Benveniste affair), but such intrusiveness and mistrust is rather rare (and a significant insult); generally lack of reproduction by others is sufficient to dismiss bad (or unlucky) work.
Cheers,
Quote
December 1, 2009, 7:54 pmzuch says:
Further re Dr. Gator’s suggested metrics for “good research”:
Moving the goalposts after you know the outcome (such as insisting on more stringent criteria for acceptance of some work) is actually one of the biggest procedural booboos in good science. You wouldn’t be doing that now, would you?
Cheers,
Quote
December 1, 2009, 8:01 pmspostrel says:
Since the CRU gang actively conspired to block publication of conflicting papers, to the extent of getting editors fired when they disapproved of their decisions, zuch’s picture of the process is misleading. Further, when a few conflicting papers were published using the data that could be accessed, the CRU gang simply said “well, you didn’t use our exact data and procedures” but wouldn’t reveal exactly what those were.
Quote
December 1, 2009, 8:02 pmzuch says:
Those scientists in general were established, with large CVs behind them and plenty of independent work of their own, rather that flacks with paid sinecures at energy-company-fronting “think tanks” and such, newly minted “climatologists” looking for their 15 minutes, or worse yet, bloggers and Internet talking heads with too much time on their hands and an antipathy to global warming research.
But for the record: CF died because no one could reproduce it. See my posts above. ‘Nuff said.
Cheers,
Quote
December 1, 2009, 8:08 pmRobert Ayers says:
Commentator Schwartz used and defined the word “robust”. That is an important point.
Small differences in coverage (temporal and spatial), as well as differences I the methods through which the data was gathered, can potentially make substantial differences in result.
If that is the case, the results are not robust. For example, if your data plus your methods produce a hockey-stick, but if (using simplistic examples) using a slightly different set of weather stations, or using the median rather than the mean of daily temperatures to derive a monthly temperature, gives completely different results, then we know that your methods are not robust and your conclusions do not follow from the data. (This was in fact shown to be the case for the Mann hockey-stick.)
Quote
December 1, 2009, 8:14 pmzuch says:
I know this well. I cited on an old GW post here the unmasking of “Sir” Cyril Burt.
No. That they didn’t have for the Burt unmasking. What they had was the results showing quite remarkable invariance despite the addition of additional data. It would be best to refrain from commenting when you don’t know sufficiently to say what you’re saying.
Cheers,
Quote
December 1, 2009, 8:16 pmzuch says:
Cite for this?
Cheers,
Quote
December 1, 2009, 8:20 pmzuch says:
... or that mean and median measures as measures of central tendency differ in fundamental ways in the system you’re measuring, implying perhaps that such are not normally distributed in the physical world you’re examining, and that there’s something else going on as well there worth looking at....
Cheers,
Quote
December 1, 2009, 8:26 pmBlue says:
As I pointed out above, since they won’t describe the source of their raw data in detail nor will they describe the methods used to transform and clean the data, there’s simply no way to show that they “got it wrong.”
Quote
December 1, 2009, 8:33 pmsteve says:
“As a general matter, the skeptics aren’t advancing the relevant claim. When one provides a theory (AGW) that person/group assumes the burden of supporting that theory. Not the reverse. “Skeptics” have it easy, they have only to find a single failure to reasonably refute the theory. ”
On this we will disagree some. Reproducibility is an important property in science. The CRU results have been reproduced by others who have been willing to what I have suggested. The skeptics should do the same. That is how we find errors, think about the Korean cloning scandal or any number of others where people could not reproduce the same results.
I am specifically thinking in two parts here. First, there is some dispute about whether there is any warming. That really should not need an exotic model, just work. Then, the skeptics should provide their own model so that they, with their better understanding of climate science, can show us what is wrong with the older models. The goal here is to get things right, not just prove the AGW folks wrong. Do you oppose the efforts of this nature?
Doc Gator-You are much more sure about this than I am. Professor Volokh has elaborated above on why data may not have been shared. I would also suggest we keep this in a proper time context. Data from 20–25 years ago, when this was not a controversial area, when the CRU was not an archival site was simply not going to kept around forever.
The truest way to determine who is correct here, is to have an open effort by all concerned. Let the best science win.
Steve
Quote
December 1, 2009, 8:40 pmHarryEagar says:
steve sez: ‘He (McIntyre) does publish his code. Strangely enough, he has not published his emails.’
He didn’t publish CRUs e-mails, either. Do you have a point?
Quote
December 1, 2009, 8:41 pmzuch says:
Perhaps you can cite a paper you take issue with, and explain where the alleged deficiencies are in their experimental protocol presentation ... and maybe some correspondence in which they refused requests for clarification might be illuminating (this means both the requests and denial).
Cheers,
Quote
December 1, 2009, 8:42 pmBlue says:
Zuch, don’t be tendentious. If they had provided that level of detail all of this discussion would be moot.
Quote
December 1, 2009, 8:46 pmElliot says:
1. If CRU threw away the raw data, then they threw away whatever they received from the national weather services.
2. The data that remains is the product of CRU processing.
3. Hence, it is impossible for them to give out what they received from the weather services, and it has been impossible for years.
4. Does anyone contend the CRU work product must be kept secret forever because it used some national weather service data as raw input data?
Quote
December 1, 2009, 8:49 pmnice strategy says:
Trillions of dollars to convert to cleaner energy is a lot of money, but it pales in comparison to the costs of actual climate change. We aren’t going to be able to prove the theory beyond all doubt before it is too late to start the transition. Are you willing to sell out future generations? Are you willing to take that chance? The CRU deal is not going to undercut all the climate science done in the last 20 years, despite longing for it to do so. Improve the model: marginal changes can change the urgency with which we need to act (in either direction) and how best to respond. The cautious thing to do is to plan for the worst. (Where’s the 1% doctrine now?)
And it isn’t like the trade-offs are simply between otherwise clean energy that gives off CO2/greenhouse gasses and clean energy that doesn’t. Coal is toxic. Burning tons of it into an atmosphere as tall as a short hike is a risk. Deny this at your peril... oh wait, it won’t really affect must of us. Deny this at the peril of people with less affluent descendants who live in places with less hospitable geography.
Quote
December 1, 2009, 8:53 pmzuch says:
For maybe the fifth time now on these GW threads, I will point out that the standard you critics have for “proof” and “evidence” of GW seems to be far stricter than that you have for the allegations of malfeasance by the researchers.
Don’t you feel even the tiniest bit guilty in screaming at the top of your lungs a demand that the CRU produce their work and results, and then refuse yourself to do the very same when asked here?
All I’m asking for is evidence for your assertions. Surely you don’t expect me to simply accept your conclusions without evidence?
Why do I have to keep asking this? Don’t people bother to read any more?
Cheers,
Quote
December 1, 2009, 8:57 pmElliot says:
Richard Lindzen of MIT is flack, newly minted, blogger, or talking head? This article by Lindzen is an excellent explanation of climate models.
Quote
December 1, 2009, 8:58 pmHarryEagar says:
zuch sez: ‘Perhaps you can cite a paper you take issue with, and explain where the alleged deficiencies are in their experimental protocol presentation ... and maybe some correspondence in which they refused requests for clarification might be illuminating (this means both the requests and denial).’
It’s been obvious for several days now that you are flying blind. Exactly what you want to see has been available at CA for years.
Years.
Quote
December 1, 2009, 9:01 pmnice strategy says:
A lot of macroeconomic policy is based on data that is incomplete, adjusted for validity in imperfect ways, and otherwise questionable.
I know work has been done to question GDP and especially the CPI, but most economists base their work on data published by governments (like the US Bureau of Labor Statistics) and I’m not sure how often the methodology for collecting the raw data and turning it into “working data” is re-evaluated from the inside or outside.
Inflation metrics are used to adjust everything to “real” dollars. If the inflation is exaggerated, growth is underestimated and benefits tagged to inflation get more generous (in _real_ real terms) over time. This could cost billions. Moreover, government policy to combat inflation that doesn’t exist comes at the cost of government policy that could be used to alleviate unemployment.
Let’s hope the econ statisticians don’t get politicized.
Quote
December 1, 2009, 9:04 pmBCKane says:
Zuch
Are you honestly suggesting that people need to be vetted by CRU or YOU before CRU should be required to release data required to actually test their assertion? I think there is a massive difference in the way you see the scientific process and the way many on this site see it.
Quote
December 1, 2009, 9:06 pmgeokstr says:
And of course that puts the skeptics in a Catch-22 of the left’s making.
Where will they get the money for this very expensive research? They won’t get it from any government, directly or indirectly, or any liberal think tank, or any of the professional organizations, who are all in on AGW.
That leaves rightwing think tanks, wealthy individuals who might actually vote Republican, or private corporations, any of which will immediately result in the predictable smear attacks as shills in the pockets of the capitalist roaders.
None of the skeptics has the money personally to fund anything like this. Even if they did, the CRU emails show that they’ll never get published anyway.
Game over.
What a wonderful fix the left has in on this issue.
Quote
December 1, 2009, 9:08 pmDuracomm says:
Zuch,
You provide a perfect example of the narrow minded tunnel vision possessed by large portions of the true believers in AGW.
You spend a huge amount of time and effort defending the accuracy of the Pro AGW science and no time considering if it is technically possible to achieve the reduction in carbon emissions that AGW advocates are demanding.
Even worse AGW supporters like yourself rarely if ever consider the possible negative unintended consequences that might be caused by policies enacted to try and reduce carbon emissions.
This one dimensional monomania resulted in the implementation of policies that have caused immense amounts of environmental destruction.
There is a certain amount of gallows humor in the fact that proponents of AGW like Al Gore and yourself have caused environmental destruction on a scale that exceeds the worst of what the oil companies have ever done.
Palm oil: the biofuel of the future driving an ecological disaster now
Much of this destruction was a direct result of environmental policies enacted to reduce carbon emissions.
Hard to say cheers to that.
Quote
December 1, 2009, 9:15 pmRichard Aubrey says:
Duracomm. No problem. Those are all brown people over there. Beneath notice.
Quote
December 1, 2009, 9:19 pmcorneille1640 says:
Maybe we should have a market solution that involves the scientists making profit-maximizing decisions, free of state intervention. That way, they won’t keep the data secret in order milk it for more publications (and prestige, and higher salaries).
Quote
December 1, 2009, 9:27 pmzuch says:
Please read. What I said is straight-forward and ought not cause you any confusion if you take care to read it.
What I said was that when Blue made these assertions:
he ought to provide the specifics and evidence to support them.
He then refused:
... dismissing my request as “tendentious”, occasioning my subsequent comment which mystifies you so.
Hope the links and chronology clear things up for you. It’s not hard to understand.
Cheers,
Quote
December 1, 2009, 9:30 pmzuch says:
Then it should be a very simple matter to round it up, right? Why the stonewall? Did no one teach you (or Blue) how to cut and paste and do links?
Cheers,
Quote
December 1, 2009, 9:34 pmzuch says:
He’s not a GW sceptic. He accepts warming is happening. He disputes the need to take action (and some of the predictive modeling).
I’m talking about the loudest majority here.
Cheers,
Quote
December 1, 2009, 9:42 pmBCKane says:
Evidently I missed these key points in your comments.
For maybe the fifth time now on these GW threads, I will point out that the standard you critics have for “proof” and “evidence” of GW seems to be far stricter than that you have for the allegations of malfeasance by the researchers.
Don’t you feel even the tiniest bit guilty in screaming at the top of your lungs a demand that the CRU produce their work and results, and then refuse yourself to do the very same when asked here?
All I’m asking for is evidence for your assertions. Surely you don’t expect me to simply accept your conclusions without evidence?
Quote
December 1, 2009, 9:45 pmzuch says:
Care to explain where I’ve ever professed (literally or de facto to be a “true believer” in AGW? Or are you just making sh*te up to bolster your ‘argument’? This gets tiring, people lying about what I’ve said.....
I’m not defending the accuracy. I truly don’t know (but am willing to give them the same benefit of the doubt generally accorded scientists). But why TF should I bother addressing the technicalities of carbon reduction feasibility?!?!? Isn’t that an entirely different issue and a subject for a different post? Or perhaps ... maybe ... in your mind, the science follows the merits of the various suggested policies, rather than the other way around.... I think I wrote a post on a previous GW thread where I said I suspected as much.
[rest of your harangue snipped as irrelevant; save it for a day where there’s a post on such, and I have put forth an actual opinon for you to attack, m’kay?]
Cheers,
Quote
December 1, 2009, 9:56 pmsteve says:
“steve sez: ‘He (McIntyre) does publish his code. Strangely enough, he has not published his emails.’
He didn’t publish CRUs e-mails, either. Do you have a point?”
Yes, so I will make it more clear, this is a law blog after all. We geeks do not treat every email as though it is a legal document. Hence, some of what was in those emails was gossip, some probably sick humor and some just venting. If we revealed everyone’s emails, what would they look like? Probably lots of fodder for complaints of hypocrisy. Immediate accusations would be made based upon one’s political affiliations. Happens all the time. Thank goodness some people like Prof Volokh here look a little deeper. Let’s publish everyone’s emails and find out what all of those scientists were hiding, even the skeptics.
“That leaves rightwing think tanks, wealthy individuals who might actually vote Republican, or private corporations, any of which will immediately result in the predictable smear attacks as shills in the pockets of the capitalist roaders.
None of the skeptics has the money personally to fund anything like this. Even if they did, the CRU emails show that they’ll never get published anyway.”
I will go out on a limb here and predict that the person(s) who definitively disprove AGW win a Nobel prize. Especially now, no one would refuse a credible effort. Funding? You jest, there are plenty who want AGW to be wrong. I dont read McIntyre, Morano or the other skeptics every day, more of a foreign policy health care guy, but the papers I remember being turned down were not the original research types. Might be good to go back and pull those out to see if they should have been published. That should be part of the investigation.
Steve
Quote
December 1, 2009, 9:58 pmguy in the veal calf office says:
The countries in which these meteorological services are located will have to, in the aggregate, spend trillions of dollars under various climate-change-fighting proposals. If I’m right that data sharing is an important part of making science accurate, those countries have much to gain from such sharing.
That seems wrong. The developing countries have consistently demanded hand-outs because of all the CO2 already emitted by developed countries, so they’d be delighted if those developed countries felt guilty enough to hand over the money.
In the developed world, many politicians and their constituents are volunteering to administer and spend new power & trillions of new money. Careers and rent-seeking company’s are built on the threat (Most of the major energy companies government intervention for rent seeking purposes.)
So, I think your view is generous and kind, but not realistic.
Quote
December 1, 2009, 10:11 pmzuch says:
I think I addressed that here. Blue seems to have complaints about the evidence and information that they did deliver (and/or its quality or sufficiency). Peer reviewers seem to have had no significant problems with this (for what that’s worth), so I asked what his problem was. What’s wrong with that?
I’ve explained my objections to many of the shriller complaints on these items — typically of at best peripheral concern to the substance of the science in question — in past threads. There’s too many comments on such to link here; please peruse past GW threads and tell me what you think is missing.
Huh? What is your point here? Thanks in advance.
Cheers,
Quote
December 1, 2009, 10:17 pmMorgan Price says:
I agree with the original post except for this part — “And to verify the CRU’s work it’s necessary to have the precise data on which the CRU relied.” Instead, as stated by some other commenters, a much more convincing reproduction would be for someone else to independently construct a record, from the individual sources, rather than to try and reproduce the exact same result as CRU. In fact, I thought that there were already other reconstructions (although I don’t know how independent they are of CRU’s).
Quote
December 1, 2009, 10:34 pmFourier says:
The Wegman report explains the McIntyre and McKitrick critique of Mann et al.
It found the critique to be “valid and compelling.” Mann and friends didn’t have a clue about the statistical techniques they were using and they misused them.
People have tried hard to replicate the Team results. A lot of discussion can be found on the ClimateAudit.com website simply by searching on replicate.
Quote
December 1, 2009, 10:50 pmDarkHelmet says:
Quoting:
“I will go out on a limb here and predict that the person(s) who definitively disprove AGW win a Nobel prize. Especially now, no one would refuse a credible effort. Funding? You jest, there are plenty who want AGW to be wrong. I dont read McIntyre, Morano or the other skeptics every day, more of a foreign policy health care guy, but the papers I remember being turned down were not the original research types. Might be good to go back and pull those out to see if they should have been published. That should be part of the investigation. ” Steve
Definitively disprove? Did you really mean to write that? If so, when did you stop beating your wife?
Once again, if you propose a theory, you are obligated to support it with evidence and reasoning if you want to convince anybody. The affirmative case has the burden of proof. The negative case prevails when it successfully challenges data, methodology, veracity, consistency or reasonableness of the affirmative case. It also has the luxury of standing silent while the affirmative case self-destructs.
Quote
December 1, 2009, 11:23 pmzuch says:
See here. Then see here.
And more from the Wiki page:
Just some useful background.
... which is of course an impartial and entirely neutral site.
Wegman was dragged into this by former Arco consultant Congressman Joe Barton, who’s ... shall we say ... a real kook (note, three different links there).
Cheers,
Quote
December 1, 2009, 11:24 pmzuch says:
They did. It’s called “publishing a paper”.
Not really. This is, to use legal parlance, “collateral attack”. The negative case prevails when it shows a different result (or one not compatible with the theory) to sufficient certainty. What you think is a “prevail[ing]” case is basically ad hominem (or as some might term it, “kvetching” or “carping”). Science tends to eschew such as being much relevant in actual science.
Cheers,
Quote
December 1, 2009, 11:31 pmJohn Moore says:
Zuch says...
Zuch says...
Zuch says...
Zuch says...
Zuch says...
Zuch says...
Zuch says...
Zuch says...
Zuch says...
Zuch says...
Zuch says...
Zuch says...
Zuch says...
Quote
December 2, 2009, 12:04 amzuch says:
John Moore:
Thanks for that substantive comment. May we see more in the future?
Cheers,
Quote
December 2, 2009, 12:06 amDoctor Gator says:
ZUCH 7:54pm
“In most cases, the standard practice is for others to reproduce the results themselves... ”
No. In this case all raw data are historical temperature measurements or tree ring data or ice core data, not lab experiments.
It is impossible for third party scientists to reproduce historical temperature measurements. They must rely on data sets created by others, usually governmental organizations who are global warming advocates (like Hadley and NASA). Some data gatherers process/adjust the raw data then record the processed data, not actual measurement, in the data set. It is known that NASA’s GISS run by James Hansen has dropped many temperature measuring stations’ data from the GISS data sets. Most dropped stations are rural. This creates a warming bias in GISS reports.
Tree ring data analysts like Mann (PSU) and Briffa (Hadley) pick and choose a few proxy trees of a species from a forest of the trees, then interpret temperature from tree ring characteristics (lots of assumptions) I think it may be impossible to reproduce this data source without the original researcher identifying the specific sample trees that must still be in the forest and/or producing the tree ring cores.
Ice core raw data is very expensive to reproduce. I don’t know that ice core producers share their physical ice cores with third party reviewers. Third party reviewers (think skeptics) can’t get grant money to duplicate ice core production next to a warmer’s core.
Quote
December 2, 2009, 12:08 amJohn Moore says:
Yes
Quote
December 2, 2009, 12:23 amjab says:
Regarding sharing of data sets:
I’m an astrophysicist... data collected with national facilities (Hubble, Chandra, etc) must eventually be shared, but lead scientists on each project usually get sole access to the data for a year... they made the proposal to point the telescope at a particular target to look at a specific phenomena, and they are given the chance to analyze and publish first. It is incredibly competitive despite the fact that the stakes are not money, but prestige... which for many academics is more important anyway.
Quote
December 2, 2009, 12:25 amKazinski says:
It took me ten posts to figure out he had no idea what he was talking about, the 50 posts after that were gratuitous.
Cheers
Quote
December 2, 2009, 12:31 amDavid Schwartz says:
You would think it would be simple. After all, people have been taking temperature readings for more than a century. Surely you can just compare the old readings with the new ones, right?
Unfortunately, the list of problems with that approach are massive.
1) How were the thermometers back then calibrated? Are they comparable to today’s thermometers? Do we have ‘jumps’ in the data from when thermometers were changed and technologies were upgraded? Can we find/fix them? Is it fair to do so?
2) Over time, new stations are added, stations are removed. How can we keep the same coverage and weighing everywhere so our results are comparable? What do we do about areas that had no coverage prior to a certain year but good coverage now? Do we “fake them” for the first years, throwing them off? Or ignore them in the later years, throwing them off?
4) How do we correct for local changes that don’t reflect global climate? What if a station was once in a rural area but is now surrounded by heat effectors.
5) What do we do about “one off” events like Pinatubo? If we count it, how do we know we’re not throwing things off by adding a rare “down” event but no rare “up” events just because none fell in the range we’re looking at? And if we include it, can we really call our result a fair sense of what the climate is doing?
Remember, we have the worst possible situation. We’re trying to compare the “old average” to the “new average” and find a small signal in the large noise. I suspect that the problem may be effectively unsolvable over the same time range as human CO2 emissions have been significantly increasing to date. We just haven’t been emitting CO2 long enough. It’s like trying to figure out if a coin is very slightly biased towards heads but only having data from 20 flips.
Quote
December 2, 2009, 12:43 amzuch says:
Correction there: It took you ten posts to figure out that you had no idea what I was talking about.
What can we do to cure that? Suggestions are welcome.
You know, I’d really not like to kill the thread with such repartee, so anyone that goes back to substance will probably be appreciated by many.
Cheers,
Quote
December 2, 2009, 12:47 amSunTzu's Nephew says:
Has the Nobel prize EVER been given for proving a fraud?
Quote
December 2, 2009, 12:56 amgda says:
zuch
You keep asking for citations, but is it just me who finds it amusing that most of your own links are to Wikipedia? As someone said “this is a law blog....”. I don’t even think Wiki is permitted as a reference source in high school, is it?
So you can’t be seriously asking that we should accept a Wikipedia entry on the Hockey Stick (or indeed on anything to do with AGW; indeed on virtually any controversial area) as a credible source?
I’ll give you the benefit of the doubt that you haven’t been following this entire “climate change” discussion over the last several years. You may therefore be unaware that William Connelly, formerly of the outed propaganda website RealClimate, rules the climate area of Wikipedia with an iron fist. It’s his interpretation of the Hockey Stick controversy which appears on Wikipedia.
Do yourself a favour and get up to speed. I highly recommend that you spend some considerable time over at ClimateAudit reading their archives. Spend some time at Lucia’s Blackboard as well. You’ll find that different opinions are permitted at those sites, unlike at RealClimate. You might note that ClimateAudit has a link to RealClimate, but that courtesy is not reciprocated.
I know only too well what it’s like to argue with a true believer, so I’m not holding my breath for you to actually take my advice. No doubt you can also find some reason to ignore Clive Crook (world-respected journalist, editor & commentator on politics & economics in such publications as the Economist and the Financial Times) who comments in The Atlantic …
gda
Quote
December 2, 2009, 1:06 amBruce Hayden says:
Another problem is that the Earth appears to have been gridded in order to provide global temperatures. But there are many potential problems with that, including that some grids apparently don’t have any temperatures, others have incomplete temperatures, etc. The solution? Essentially interpolations. Missing right now is how they got from point A (actual temperature readings — and which ones they used) to point B (gridded temperatures over the requisite time frame).
Quote
December 2, 2009, 1:14 amzuch says:
Care to explain how you’re contradicting me?
Which are not available [historical temps] or can’t be remeasured [tree rings, etc.]? You know, if I didn’t trust the CRU with a passion, I’d hope I had the good sense to avoid asking them for data — nay insisting they turn it over and calling them frauds if they didn’t — and instead try and get it from trustworthy sources....
CRU didn’t do the ice cores. If your claim is indeed true, this is CRU’s problem exactly how? And if CRU had to pay to get it, why should anyone else get a free ride (if it is even legal for CRU to give the data out for free)?
So do the right thing, and at the very least, double the sample size ... or provide some real data that are not polluted by those frauds. Everyone will thank you.
Here’s my money comment on the compulsive “need” here to “reproduce” with the exact same old data. If you don’t understand it, you don’t have any real (that is, tempered by the real world) idea of what science is all about.
Then again, I suspect the “compulsion” is simply an attempt to demand the impossible (or at least unreasonable and/or time and resource consuming), and then say when the demands aren’t met with sufficient obeisance, that the whole edifice is rotten.... No. No honest person would do that kind of stuff....
Cheers,
Quote
December 2, 2009, 1:26 amzuch says:
I don’t see why not. Cut’n’pasting, of course, is frowned on. But I’d point out that Wikipedia does a pretty good job of providing further links ... something noticabl absent here amonst most (although I try to do my part).
But to get to the nub of the matter, if someone fulfilled my request with a cite to Wiki, I wouldn’t complain per se (assuming that the cite actually answered my request).
It gives some context on the controversy. I thought it might be illuminating information. Whether you believe it or not is up to you; I have no supernatural abilities there to control your perceptual faculties. And just a FYI: Down at the bottom of the Wiki page are further links to source material.....
Are you assuming they’re “up to speed”? Wouldn’t that be assuming your conclusions?
See here (towards the top, in case you need a hint).
But ... just OOC: Do you have a mirror in your house? I’m beginning to wonder (if by chance you feel a faint burning sensation, yes, that was the intent).
Cheers,
Quote
December 2, 2009, 1:43 amDavid Schwartz says:
That is another way of describing what I’m talking about in problem 2.
Gridding is CRU’s solution to that problem. It’s a solution with advantages and disadvantages. The big advantage is that if you get it right, your data is the most maximally useful data set you could have. You can simply average each grid over a forecast year and get one number “global average temperature” for that year. It solves the weighing problem — each grid gets the same weight, regardless of how many stations are in it.
However, it has a lot of problems as well. The biggest problem is that your output will contain data of varying quality. Some grids for some years will contain lots of good data from lots of nearby stations. Some grids for some years will contain no good data from any station and will purely be “faked”. (By “faked” I mean your best guess, but the error bars are much higher.)
The biggest disadvantage of gridding is that as time moves forward, the data gets better. So you have already a progressive “bias” going from old data to new data. If that bias at all “leaks” into a temperature bias, your long-term trend will be off. And it’s precisely a long-term trend that we’re looking for.
Suppose you have some complete nonsense readings. Maybe the station code was typed in wrong. Maybe the sign was blow or someone entered data in the wrong units. You need a “sanity check” level to reject invalid data, and we know CRU had such things.
The more stations you have in a grid, the more likely you can reject bad data. The data has to fall in a narrower range to be considered valid if you have a better idea of what the reading should be.
So as the years go on, you will be able to reject a higher percentage of the bad data. But perhaps improvements in technology and automation will mean a lower percentage of the data is bad in the first place. So over the years you may have a trend towards less bad data in the results or you may have a trend towards more. Or you may have both for different types of bad data. (We know how station locations were identified changed, so those errors may have gone down or up as other errors moved in different directions.)
Now, some errors may consistently bias data in a different direction. For example, if a station takes a reading at a particular time, and a cloud happens to shade it, that may result in a downward error. If a minus sign is omitted from a near-zero temperature, that may result in an upward error.
So the various changes in rates of the various types of errors in the input and the changes in their rates of error getting into the output may add a variety of downward and upward pressures to the result data set.
Because gridding means that a lot of your data will have to be synthesized, it will tend to magnify the effects of such errors.
Quote
December 2, 2009, 1:52 amMalvolio says:
A huge disadvantage is that the grids are all different sizes! Each grid represent a five-degree “square” of latitude and longitude. A grid near the equator is over 100,000 square miles, one touching a pole is barely 5000. (I was going to include my calculations but in the spirit of day, I’m going to resist every attempt to get me to explain it and then let a hacker steal reams of my indecipherable FORTRAN code on the subject.)
I’m guessing there are also a lot more measuring stations along the 25,000-mile equator than on the 350 miles of the 85°N meridian. Indeed I would be surprised if there were more than a handful of stations in 72 grids comprising the northern-most band (120,00 square miles of sea-ice and open water around the North Pole), but their data is as influential as all the stations with 175 miles of the equator (8,750,000 square miles and a lot more of it dry, warm land).
Do I know that CRU doesn’t weight the grids by area? I don’t know for sure, but these are people who can’t calculate great-circle distances.
Quote
December 2, 2009, 3:46 amDavid Schwartz says:
Malvolio: CRU publishes their final grids. So anyone can weight them themselves. So they couldn’t have made that mistake.
But any approach can be done wrong, so I don’t think it’s fair to say the different sizes of the grids is a disadvantage to gridding. Had I made the decision, I believe I would have chosen gridding too. (Though my data processing methodology would have been very different from theirs, I think.)
Quote
December 2, 2009, 5:26 amwb says:
When research codes are tens of millions, and computer time requirements on large capacity machines are very large, this is not very possible. The fact is that governments can stack the deck.
Consider the case of high energy physics, doing one’s own experiment is not possible. That is why competing teams get funded and why data sets are available to outside groups, one the box of data is opened. Typically groups do considerable blind or better double blind analysis before “opening the box.”
As to when a group has an obligation to release raw data. That greatly depends on whether selective data adjustment is made. If one cannot compared raw with adjusted sets a referee or skeptic has no way to know if the data was handled in a scientifically ethical fashion.
I was at dinner last night with a dozen internationally known physicists of a wide range of political persuasion and opinion about AGW and what to do about increasing CO2 levels. All were distressed at the grave damage to the credibility of scientists done by the CRU group.
Quote
December 2, 2009, 5:58 amA. Zarkov says:
If he deletes his raw data, he’s a fool. There is absolute no reason to “toss” your original data. Data storage is extremely cheap these days. Note he says he does keep the old and somewhat bulky digital magnetic tapes in his office. Why would anyone keep the tapes, which he doubts he can read, and “toss” the data from the tables which he should have put on DVDs, or have resident somewhere on a hard disk. BTW you can buy a 1 TB (that’s 1,000 giga bytes) hard disk for $180, and this is an expensive source for hard drives. Moreover one should note he hasn’t “tossed” his data, he’s kept it all including the bulky tapes. Scientists are pack rats, they don’t toss anything. This whole post is designed to try to defend the indefensible for political reasons– it’s pure propaganda.
Quote
December 2, 2009, 6:25 amsteve says:
How old are you? We are talking data from the 80s here. He kept some mag tapes, he does not say all, as he notes he deletes stuff that is of no use. Please note that some of the CRU data being discussed was on paper. The guy is telling you what he does, and I presume others do also. This would have made a lot of sense prior to the total digital age we have now. You really needed archival sites.
WB-But this is not running new experiments. This would be collecting old data and running your own models. You dont need to build a collider. Heck, they could just use existing models and correct them where they think they are wrong.
I would agree that this has done damage to the credibility of the CRU group and AGW in general. However, I am much more interested in knowing if AGW is a problem. Other researchers using other data have come to similar conclusions. Suppose, arguendo, that there really is a global conspiracy including the folks at NASA. Suppose they had secret communications with the people at CRU so that it does not show up in the emails. The best way to be sure is present different results in an open, transparent manner.
You do realize part of the problem is just honest differences of opinion on what data to use and what it means? Surely that would be understood on a law blog?
Steve
Quote
December 2, 2009, 7:17 amA. Zarkov says:
It’s pretty obvious he deletes pretty much nothing he can’t easily recreate. Scientists are pack rats. Look at their offices in places like Los Alamos, Livermore, Oakridge, Brookhaven, Watson, Bell, Sandia. The older guys have card decks, punched paper tape, print outs, microfilm, microfiche, floppies, digital mag tape, and now days CDs and DVDs– clutter everywhere. I have been in hundreds of offices in a lot of the above places.
I simply don’t believe CRU threw away their original data. They were either careless and lost it, or hid it somewhere to avoid a FOI request. Unless a lot of people are willing to lie under oath, this will eventually come out. Of course we might not get an honest investigation for many years because AGW is a big honey pot for a lot of interests.
Quote
December 2, 2009, 7:42 amzuch says:
Quote
December 2, 2009, 7:42 amDavid Schwartz says:
The problem is that many of the skeptics believe that doing what CRU claimed to have done is impossible (or at least that they strongly doubt there is any way to do it). So how are they supposed to replicate what CRU did? Trying and failing doesn’t prove impossibility.
The only people who are able to reach such conclusions are the people who believe we have enough information to reach conclusions. You will not see competing conclusions from the other group. What you will see is a response to each claimed conclusion that it cannot be right because of its various (unfixable) flaws. (And probably some fixable flaws, we’re only human after all.)
You can jump to my comment here and here to see some very simple explanations of some of the reasons why what CRU claimed to have done cannot be done.
Though the last comment is specific to gridding approaches, and gridding is not the only approach, the other approaches are actually worse. CRU was right to choose gridding.
Quote
December 2, 2009, 7:43 amA. Zarkov says:
They are working from essentially the same data sets as there is tremendous overlap. Whatever biases have been introduced with “adjustments” will propagate. Then there is the group-think effect. At this point I’m not sure we have any credible evidence of warming at all.
Look the the earth’s temperature is a four-dimensional field. Latitude, longitude, altitude and time. This field is very sparsely sampled in both space and time. We almost no data from the southern hemisphere. From this sparse sampling we a suppose to detect about a 1 degree C change over a century! Very hard to do, if not impossible because the measurements are noisy. Even if we can establish a temperature increase, the cause could easily be natural. This whole business has little to do with science. Follow the money, it’s all about money.
Quote
December 2, 2009, 7:54 amdearieme says:
There’s little point people still claiming that all AGW “science” is peer-reviewed when the e-mails reveal that instead it is often crony-reviewed. There’s no point referring to “valid” this-that-and-the-next-thing when validity is what is in contention.
Quote
December 2, 2009, 9:19 amalkali says:
Data storage is extremely cheap these days. Note he says he does keep the old and somewhat bulky digital magnetic tapes in his office. Why would anyone keep the tapes, which he doubts he can read, and “toss” the data from the tables which he should have put on DVDs, or have resident somewhere on a hard disk. BTW you can buy a 1 TB (that’s 1,000 giga bytes) hard disk for $180, and this is an expensive source for hard drives.
My understanding is that to the extent material was discarded, that took place in the 1980s, when storage was bulkier and more expensive than it is now.
Quote
December 2, 2009, 9:24 amMark Buehner says:
There is a burden of proof issue here.
Moreover– if CRU did their work correctly it would be easy. CRU would put a link to the raw data, the dataset, and the metadata explaining what was ‘value added’ to the data and why, their code, and when you run the code with that dataset the results will match.
The attitude of the processor is immaterial. Now of course a skeptic could look at the assumption made in the dataset and question it, and that’s fine. If they make a point that is clearly important, enough people (fellow scientists surely) will ask the question until it gets addressed. Or somebody could make that change and run the program again and see if the answer changed materially. People can also look at the code and find out if there is any hand-waving and outright kludging of the type we saw in Harry_read_me.
Quote
December 2, 2009, 9:45 amalkali says:
@EV: And to verify the CRU’s work it’s necessary to have the precise data on which the CRU relied, and not just other data from other sources that has its own limitations and characteristics. Small differences in coverage (temporal and spatial), as well as differences in the methods through which the data was gathered, can potentially make substantial differences in result. And absence of the precise data, with its precise temporal and spatial boundaries, makes it impossible to verify the particular results that CRU reports. If you are so important, and your dataset is so important, it’s important that other scientists who want to check your work have access to your dataset and not just to other datasets.
Most of the data used by CRU is publicly available. The nuances of the particular dataset used (e.g., whether you happen to have data for a particular location) shouldn’t matter if the result is robust.
Stepping back a bit, it is not correct that as a general matter of modern scientific practice, scientists make their raw data and computer code used to process that data available to anyone who asks. Maybe that should be the practice, but it is not now the practice. Changing to that practice would impose substantial (although perhaps justified) burdens on researchers.
By way of comparison, if someone asked Prof. Volokh for all the drafts and work materials he used or reviewed in writing his most recent law review article, he would probably decline to respond to that request. Would it be of some value to see all those things, so that we could be scrupulous in determining whether Prof. Volokh was being perfectly intellectually honest at all times? Well, perhaps. But the burden on Prof. Volokh of maintaining those records, and of answering questions about them months or years later, would be unreasonably high. If we want to challenge his conclusions, we should gather our own source material — including by looking to the source materials referenced in his published article — and do our own work.
One interesting example of the burden that would be associated with a more demanding standard of scientific transparency: One commenter has pointed to computer code attached to one of the hacked e-mails that applies a “VERY ARTIFICIAL” adjustment to the dataset — the quote comes from the comments to that code — and has argued that this shows that the results published by the author of that code cannot be relied upon. A second commenter looking at the same code noticed that the portion of the code that actually makes that adjustment had been “commented out” (i.e., marked to be inoperative). The original commenter nevertheless argued that at one point the adjustment must have been used in some presentation at some point and suggested that the author of the code should be required to account for that. The impracticality of requiring that level of transparency seems obvious to me.
Quote
December 2, 2009, 9:47 amMark Buehner says:
I stopped reading there because that is a ridiculous assertion. The dataset is critical. It includes which stations where used (not all of them by far are used to create a dataset, rather important to know if you want to replicate CRU’s work), how the gridding was done, and what was done to the data of each set to correct for all sorts of things that must be corrected for. This is critical to understanding this scandal.
A dataset like HADCRUT3 will have thousands of value added changes to the original data to reflect things like stations that moved, adjusting for seasonal changes like El Nino, all sorts of things. Just what was done to the raw data and why is absolutely crucial. And apparently, that information no longer exists (at least not in any format that is reproducible). That calls into question all the studies done with that dataset, which is a large number of very influential studies.
Quote
December 2, 2009, 9:58 amRichard Aubrey says:
alkali.
Ref yr. last graf.
That it is impractical to be that transparent does not affect the conclusions drawn about the effect of the possible discrepancy.
Quote
December 2, 2009, 10:05 amMark Buehner says:
And lets not confuse impracticality with inconvenience. The idea that CRU never considered that they could be called upon to reproduce their work or that a 3rd party might wish to do it says more about the current state of climate science than anything else.
Quote
December 2, 2009, 10:11 amDuracomm says:
Zuch says,
1. Because, as the example I provided shows, not addressing feasibility results in policies that cause massive environmental destruction and increased carbon emissions.
2. If carbon reduction is not feasible policies regarding AGW must switch from Carbon reduction to mitigation of changes caused by AGW.
This will have an enormous impact on how resources are allocated to respond to AGW.
In other words you should care about the feasibility of carbon reduction because it is vastly more important than what the latest temperature reconstruction or climate model shows.
Quote
December 2, 2009, 10:14 amalkali says:
I wrote: Most of the data used by CRU is publicly available. The nuances of the particular dataset used (e.g., whether you happen to have data for a particular location) shouldn’t matter if the result is robust.
@Mark Buehner: I stopped reading there because that is a ridiculous assertion. The dataset is critical.
Ridiculous? I don’t think it’s even particularly controversial. It’s sort of a basic premise of science that the correctness of a result doesn’t depend on the particular observations made. You don’t need Newton’s data about the speed of an block on a inclined plane to determine whether his results are correct.
A dataset like HADCRUT3 will have thousands of value added changes to the original data to reflect things like stations that moved, adjusting for seasonal changes like El Nino, all sorts of things. Just what was done to the raw data and why is absolutely crucial. And apparently, that information no longer exists (at least not in any format that is reproducible). That calls into question all the studies done with that dataset, which is a large number of very influential studies.
A copy of HadCRUT3 and lots of explanatory information can be found on this page.
Quote
December 2, 2009, 10:17 amMark Buehner says:
Pointing to a huge pile of data that you took pieces of and manipulated and then fed through a computer code and then saying– ‘there, your answer is in there somewhere’ is not making your experiment reproduce-able.
Not remotely equivalent. It would be more like Einstein pointing to a 100 years of astronomical data and saying ‘everything you need to know about relativity is right there’.
We’ve been all over this– none of the intermediate data and metacode is available, and I know because that was what the FOIA request asked for. It was like pulling teeth just to get the station list out of CRU, and they refused to release the data associated with each.
If you can show me where i can find a way to relate HADCRUT3 back to the GISS raw data, congrats, because nobody else on the planet has been able to find that info including CRU.
Quote
December 2, 2009, 10:38 amzuch says:
Oh, I care about it. I just think it’s a bit OT in this thread.
And I do get a little annoyed that comments on the costs of certain policies (w/o any evidence of how those costs were derived, I might add) keep creeping in to the the discussion of the existence of GW, as if it has some relevance to the truth value thereof.
It is true that if mitigation of any putative AGW is impossible for practical or financial reasons, the question of GW becomes just a tad more ‘academic’ ... but I’d say that it’s clearly not a very ‘academic’ discussion, no? If you think that the suggested policies are impossible or too expensive, out with your work and argue that. But probably better elsewhere....
Cheers,
Quote
December 2, 2009, 11:35 amzuch says:
OK, you got your scalps. Jones has stepped down as CRU director, and Mann is facing an “inquiry” from Penn State (not to mention a probable Congressional subpoena if the eedjit Inhofe gets his way; McCarthy move on over...).
Nice Swiftboating. Well done. Congratulations.
Cheers,
Quote
December 2, 2009, 11:48 amJed Rothwell says:
Zuch wrote:
“But for the record: CF died because no one could reproduce it.”
That is incorrect. Cold fusion was reproduced by thousands of scientists in hundreds of labs. The results are often at very high sigma, and the reproducibility is no longer a problem. The NRL recently reported they were able to trigger the reaction hundreds of times in a row without a failure.
I have a collection of 1,200 peer-reviewed journal papers on cold fusion copied from the library at Los Alamos, plus 2,500 others from proceedings national and corporate laboratories. I suggest you review this literature before commenting on the research. See:
http://lenr-canr.org
Quote
December 2, 2009, 11:48 amCraig Goodrich says:
How much longer will people keep pretending that there ever has been any evidence for anthropogenic global warming? There never was any, even in the most tendentiously distorted papers from the Hockey Team. Twenty years and a hundred billion dollars later there still is no confirming evidence, and in fact the fancy satellites and buoys some of our tax money has bought provides more disconfirmation with every telemetric download.
Has anyone ever actually read, for example, Chapter 9 of WG 1 of the IPCC’s Fourth Assessment? It pretends to provide “attribution” for the 1975–1998 temperature rise (if any), but if you study it carefully you find exactly the same evasion used in 1988 — “We can’t account for the warming in our computer models, you see, unless we stuff in something about CO2.” This is idiotic; I’ve been writing computer programs for forty years and could write a financial model — “I can’t account for the Crash of 08, you see, unless I include the parameter for elephant flatulence.”
Jim Hansen is an astronomer who was badly traumatized by studies of Venus early in his career. The trauma obviously unbalanced his mind. All of the IPCC’s “climate models” are hackings of his original model designed to simulate Venus, observed from a hundred million miles away, and to get some idea of the probable quality of the hacking, see HARRY_READ_ME.txt in the FOIA archive.
This whole mess is a fraud, based on a wildly improbable theory that would have seemed silly to anyone who had passed 7th grade Earth Science back when our schools still actually taught anything. And we are still squandering billions on this nonsense while turning precious wilderness and countryside into wind wastelands in some feckless search for perpetual motion.
Scotty, beam me up...
Quote
December 2, 2009, 11:53 amMark Buehner says:
Swiftboating: To hold your betters accountable for their actions.
Quote
December 2, 2009, 11:57 amRichard Aubrey says:
Zuch’s been exhibiting the flailing of somebody back on his heels and not able to regain the initiative for some time.
To call these “scalps” is absurd.
These guys screwed up and are being investigated.
zuch implies that not investigating them would be the right thing to do, which is about all you need to know about his position.
Quote
December 2, 2009, 12:11 pmAllen says:
I really would like to see some research on the quality and techniques used in the software they wrote. I’ve been hounding a couple of tech blogs I read to do a writeup reviewing the quality but I only get back replies that the software is fine and that they’re researchers, they don’t have time or resources to invest in quality code like a business. I’ve got no idea why anyone thinks a business that produces widgets has more resources to devote to writing software than researchers who produce software models.
I know this is a lawblog and not the place to find technical analysis of software, but I don’t have any ins with the techblogs to ask for analysis, and I don’t have history with Fortran (just VBA) to do real analysis myself. You folks have any suggestions?
Quote
December 2, 2009, 12:45 pmMark Buehner says:
Here you go Allen. It’s not pretty.
Quote
December 2, 2009, 1:03 pmJohn Moore says:
Actually, the data doesn’t seem to be getting better with time. In fact, in a paper I just read, the number of stations available is going down rapidly.
Likewise, the instrumentation isn’t better — thermometry has been well known for a long time. The instrumentation placement is a big problem, and there are tons of problems, from documented biases (e.g. official stations in the exhaust of air conditioner condensers) to microclimate issues, to distorted records.
Fundamentally, near-surface temperature measurements are a terribly way to determine the earth’s temperature! Satellites offer much better hope (and fail to confirm the hockey stick).
Quote
December 2, 2009, 1:08 pmHarryEagar says:
Satellites also offer global coverage, which surface sensors have not.
They are better on the daytime side of the globe than the nighttime side, though, so even with satellites the precision is less than we would wish.
Quote
December 2, 2009, 5:28 pmTweets that mention The Volokh Conspiracy » Blog Archive » More on Data Sharing and Climate Change Research -- Topsy.com says:
[...] This post was mentioned on Twitter by PostRank – Law, Eugene Volokh. Eugene Volokh said: More on Data Sharing and Climate Change Research: I much appreciate all the responses to my earlier question ab.. http://bit.ly/7D6OZc [...]
zuch says:
Wow. Problem solved; we won’t have to burn any fossil fuels (with the additional thermodynamic advantage to fusion at room temperatures). Sing hallelujah, everybody!
Egg on the face of all those nasty JIR folks, handing Pons and Fleischmann that Ig Nobel.
Cheers,
Quote
December 2, 2009, 10:35 pmDavid Schwartz says:
For what it’s worth, cold fusion is 100% real and reproducible.
http://en.wikipedia.org/wiki/Muon-catalyzed_fusion
Sadly, it does not look at all promising as a power-generation mechanism.
Quote
December 2, 2009, 10:51 pmDarkHelmet says:
Zuch said:
The negative case prevails when it shows a different result (or one not compatible with the theory) to sufficient certainty.What you think is a “prevail[ing]” case is basically ad hominem (or as some might term it, “kvetching” or “carping”). Science tends to eschew such as being much relevant in actual science.Cheers,
end Zuch
Well, excuse my “kvetching and carping” (nice ad hominem there, btw) but this is simply wrong. Say Scientist X does some research, develops a theory and comes up with data and lines of reasoning that he claims support his theory. Yes, Skeptic Y could develop a new data set, run it through X’s methodology and come up with a different answer that would constitute a challenge to X’s theory. But Y does not need to do so. He can just as effectively challenge X’s theory by showing:
a) X’s data is incorrect, incomplete or fraudulent or of low reliability
or
b) X’s chain of reasoning has serious flaws
In the case of CRU it seems that both a) and b) have been detected. The theory is therefore successfully challenged. AGW may still be true, but CRU’s work does not make the case.
But you know all this and are simply wasting everyone’s time for reasons of your own.
Quote
December 2, 2009, 11:26 pmDavid Schwartz says:
And in any event, we can’t do that in this case, because nobody knows what CRU did to the data. They applied a large number of undocumented ad-hoc “fixes”, and we maintain there is no right way to “fix” it. So whatever data we produced would be useless. If it failed to replicate CRU’s, it would not call CRU’s into question because we admit we don’t know how to apply the right “fixes”.
See my other post about the numerous unsolvable problems in attempting to replicate what CRU did with any historical set of land temperature station data. There is just too much noise and too little data to find a signal as small as is suggested.
I’m sorry, good enough records just do not exist. The past cannot be reconstructed to that level of detail by any mechanism known today.
Quote
December 3, 2009, 3:28 amzuch says:
No. The asymptote here is simply a raw theory and no (good) data. You can’t say the theory’s wrong because there’s no data (but OTOH you can’t say it’s right). I think “challeng[ing a] theory” scientifically consists of more that just saying “There’s no data either way; I don’t believe it”. Much stronger to say, “this data shows it cannot be true.”
This hasn’t been done.
Cheers,
Quote
December 3, 2009, 6:07 amWinston says:
Irrespective of any data access restrictions at the University of East Anglia’s Climatic Research Unit, researchers should find open access to federal and federally-financed research in the US.
The Obama Administration’s new Open Government Directive creates an unprecedented commitment to agency transparency — including release of data. The Directive requires agencies to create a “strategic action plan for transparency” that includes identification of “high value information not yet available and establishes a reasonable timeline for publication....”
The Directive also includes an emphasis on ensuring that “information conforms to OMB guidance on information quality” as required by the Data Quality Act.
The Directive’s transparency requirements are in addition to those created by the Data Access Act which requires agencies to “ensure that all data produced under” federal award be made publicly available. OMB subsequently limited the access requirement to data used “in developing an agency action that has the force...” of law. Any climate change regulations or information collections would likely meet this test.
The triple hammer of the Open Government Directive, Data Quality Act, and Data Access Act establish mechanisms for providing public access to the data and models underlying federal climate change policies, as well as a process for ensuring the quality and reliability of those data and models.
It’s now up to the public to make sure that the federal transparency, access and quality mechanisms are actively used.
Please see http://www.thecre.com/wdw/2009/20091221.html
Quote
December 21, 2009, 4:03 pmWinston says:
Irrespective of any data access restrictions at the University of East Anglia’s Climatic Research Unit, researchers should find open access to federal and federally-financed research in the US.
The Obama Administration’s new Open Government Directive creates an unprecedented commitment to agency transparency — including release of data. The Directive requires agencies to create a “strategic action plan for transparency” that includes identification of “high value information not yet available and establishes a reasonable timeline for publication....”
The Directive also includes an emphasis on ensuring that “information conforms to OMB guidance on information quality” as required by the Data Quality Act.
The Directive’s transparency requirements are in addition to those created by the Data Access Act which requires agencies to “ensure that all data produced under” federal award be made publicly available. OMB subsequently limited the access requirement to data used “in developing an agency action that has the force...” of law. Any climate change regulations or information collections would likely meet this test.
The triple hammer of the Open Government Directive, Data Quality Act, and Data Access Act establish mechanisms for providing public access to the data and models underlying federal climate change policies, as well as a process for ensuring the quality and reliability of those data and models.
It’s now up to the public to make sure that the federal transparency, access and quality mechanisms are actively used.
Please see http://www.thecre.com/wdw/2009/20091221.html
Quote
December 21, 2009, 4:20 pm