Notices
Results 1 to 38 of 38

Thread: House of Commons Report on the CRU data hack

  1. #1 House of Commons Report on the CRU data hack 
    Forum Isotope Bunbury's Avatar
    Join Date
    Sep 2007
    Location
    Colorado
    Posts
    2,590
    The House Science and Technology Committee has released its report. The report is focused on academic integrity, not on the science, which other reports will cover.

    http://www.publications.parliament.u...h/387/387i.pdf

    A few snippets:

    136. Conclusion 1 The focus on Professor Jones and CRU has been largely misplaced.

    137. Conclusion 2 In addition, insofar as we have been able to consider accusations of dishonesty—for example, Professor Jones’s alleged attempt to “hide the decline”— we consider that there is no case to answer.
    Even if the data that CRU used were not publicly available—which they mostly are— or the methods not published—which they have been—its published results would still be credible: the results from CRU agree with those drawn from other international data sets; in other words, the analyses have been repeated and the conclusions have been verified. (Paragraph 51)
    Critics of CRU have suggested that Professor Jones’s use of the word “trick” is evidence that he was part of a conspiracy to hide evidence that did not fit his view that recent global warming is predominately caused by human activity. The balance of evidence patently fails to support this view. It appears to be a colloquialism for a “neat” method of handling data. (Paragraph 60)

    Critics of CRU have suggested that Professor Jones’s use of the words “hide the decline” is evidence that he was part of a conspiracy to hide evidence that did not fit his view that recent global warming is predominantly caused by human activity. That he has published papers—including a paper in Nature—dealing with this aspect of the science clearly refutes this allegation. In our view, it was shorthand for the practice of discarding data known to be erroneous. We expect that this is a matter the Scientific Appraisal Panel will address. (Paragraph 66)

    The evidence that we have seen does not suggest that Professor Jones was trying to subvert the peer review process. Academics should not be criticised for making informal comments on academic papers. The Independent Climate Change Email Review should look in detail at all of these claims. (Paragraph 73)


    Reply With Quote  
     

  2.  
     

  3. #2  
    Forum Professor Wild Cobra's Avatar
    Join Date
    Sep 2009
    Posts
    1,140
    Thank-you for the link. I haven't read but what you linked. I am disappointed you didn't include the complete context of the findings however. Hope you were just summarizing rather than cherry picking.
    135. Consideration of the complaints and accusations made against CRU has led us to three broad conclusions.

    136. Conclusion 1

    The focus on Professor Jones and CRU has been largely misplaced. On the accusations relating to Professor Jones’s refusal to share raw data and computer codes, we consider that his actions were in line with common practice in the climate science community. We have suggested that the community consider becoming more transparent by publishing raw data and detailed methodologies. On accusations relating to Freedom of Information, we consider that much of the responsibility should lie with UEA, not CRU.
    Spot on considering these people would have a limited understanding on such sciences...

    1) They note that the lack of sharing with the public is common with the AGW crowd.

    2) They suggest transparency.
    137. Conclusion 2

    In addition, insofar as we have been able to consider accusations of dishonesty—for example, Professor Jones’s alleged attempt to “hide the decline”— we consider that there is no case to answer. Within our limited inquiry and the evidence we took, the scientific reputation of Professor Jones and CRU remains intact. We have found no reason in this unfortunate episode to challenge the scientific consensus as expressed by Professor Beddington, that “global warming is happening [and] that it is induced by human activity”. It was not our purpose to examine, nor did we seek evidence on, the science produced by CRU. It will be for the Scientific Appraisal Panel to look in detail into all the evidence to determine whether or not the consensus view remains valid.
    Yep, innocent till proven guilty. Understandable with their limited inquiry.

    I wonder if there will be a Scientific Appraisal Panel looking into this.
    138. Conclusion 3

    A great responsibility rests on the shoulders of climate science: to provide the planet’s decision makers with the knowledge they need to secure our future. The challenge that this poses is extensive and some of these decisions risk our standard of living. When the prices to pay are so large, the knowledge on which these kinds of decisions are taken had better be right. The science must be irreproachable.
    Not only must the science be irreproachable, but it must be transparent. Until the AGW crowd allows the populous at large to see their data and methodology, it will never be irreproachable.


    Reply With Quote  
     

  4. #3  
    Moderator Moderator
    Join Date
    Apr 2007
    Location
    Washington State
    Posts
    8,416
    "Not only must the science be irreproachable, but it must be transparent. Until the AGW crowd allows the populous at large to see their data and methodology, it will never be irreproachable."

    That is neither how science works, nor how it should ever work. We passed the point of the general population understanding the raw data and analysis methodologies associated with specific scientific fields several centuries ago. "Irreproachable," which is tantamount to "perfect," is an unrealistic and altogether unnecessary standard for nearly any human endeavor, whether that be conducting science or deciding the fate of a murder case. The methods of communicating science conclusions to the general public incapable of understanding the detail (due to time, education etc) are the important parts here--not the unrealistic quest for perfection.
    Meteorologist/Naturalist & Retired Soldier
    “The Holy Land is everywhere” Black Elk
    Reply With Quote  
     

  5. #4  
    Forum Isotope Bunbury's Avatar
    Join Date
    Sep 2007
    Location
    Colorado
    Posts
    2,590
    Not only must the science be irreproachable
    I also had trouble with this statement, but then consider that the report is on academic integrity, rather than the science itself, and the intent of the statemtn is not to expect perfection in the science but to expect integrity, with a goal of objectivity in the process.
    Reply With Quote  
     

  6. #5  
    Forum Professor Wild Cobra's Avatar
    Join Date
    Sep 2009
    Posts
    1,140
    As I read the report, I think the things that bothers me the most are that only 5 of the 14 members were present. Hardly a quorum to decide anything important. Then on top of that, there was a consistent 3 to 1 split on voted decisions. Seems more like a white-wash to me.
    Reply With Quote  
     

  7. #6  
    Moderator Moderator
    Join Date
    Apr 2007
    Location
    Washington State
    Posts
    8,416
    Quote Originally Posted by Bunbury
    Not only must the science be irreproachable
    I also had trouble with this statement, but then consider that the report is on academic integrity, rather than the science itself, and the intent of the statemtn is not to expect perfection in the science but to expect integrity, with a goal of objectivity in the process.
    However, their recommendation are inconsistent or at least ill conceived. You don't insure integrity of the process by releasing raw data and methodology to the masses who don't have the capacity to understand it, put it in context or really evaluate it's contents while providing fodder for those deliberately spreading misinformation to confuse the ignorant public--you do it by providing independent oversight processes--as it done for economic forecasting, medical drug testing and just about every other science who's results might harm individuals or a component of the society. Arguably atmospheric science, primarily because it's mostly public funded, is already about the most open among the sciences with most data available in near real time and science publications or government technical reports available in any respectable library. Try to do that for GM crop testing for example...good luck.
    Meteorologist/Naturalist & Retired Soldier
    “The Holy Land is everywhere” Black Elk
    Reply With Quote  
     

  8. #7  
    Suspended
    Join Date
    Sep 2009
    Location
    New Orleans
    Posts
    1,191
    So transparency of data an methodology leads to confusion? That's an interesting theory.
    Reply With Quote  
     

  9. #8  
    Forum Professor
    Join Date
    Sep 2007
    Posts
    1,079
    Should all data, everywhere, be published? Al medical data ever generated in any clinical study anywhere, all biochemical data ever generated anywhere in any study, and so on? All data? All of it?

    Should all replicates of every experiment be reported? All data that perhaps was thrown out due to an aberrant factor such as a contaminant or use of the wrong controls, should that data be released as well?

    Should data from all universities be released? Data from lesser institutions? What about the data that summer interns generate? What about data that year 10s generate in their classes? Data that children generate as they are learning science, should that be released?

    Should anyone that wishes to import their own temperature measurements from their gardens, should that be incorporated too? individual records of rainfall and birdsong and so on, should this all be part of the public domain?

    How long should data be available? Should data from the 1500s be easilty retrieved? Data that the world is flat, or that disease is spiritual, or that witches and demons exist, should this data be recorded?
    Reply With Quote  
     

  10. #9  
    Suspended
    Join Date
    Sep 2009
    Location
    New Orleans
    Posts
    1,191
    I would think it is appropriate to release just the data and methods required to allow for a completely independent review,analysis and verification of results and conclusions, if one wants those conclusions to be seriously considered, particularly when the conclusions can have a large impact on public policy or society norms. Other than that, or if you don't care to have conclusions taken seriously, no need to release data.
    Reply With Quote  
     

  11. #10  
    Forum Isotope Bunbury's Avatar
    Join Date
    Sep 2007
    Location
    Colorado
    Posts
    2,590
    You don't insure integrity of the process by releasing raw data and methodology to the masses who don't have the capacity to understand it, put it in context or really evaluate it's contents while providing fodder for those deliberately spreading misinformation to confuse the ignorant public
    No you certainly don't, but when the work has been publicly funded it's apparently the law in the UK that the data should be available to anyone who requests it. Releasing it doesn't ensure integrity but witholding it could create the impression of a lack of integrity. I completely understand Jones' anger and frustration at the repeated spurious requests from people who had no real interest in doing science, but perhaps dealing with such people goes with the territory.
    Reply With Quote  
     

  12. #11  
    Forum Professor Wild Cobra's Avatar
    Join Date
    Sep 2009
    Posts
    1,140
    There was a suggestion to simply have the data available, like NASA does. What would the harm be in that?
    Reply With Quote  
     

  13. #12  
    Moderator Moderator
    Join Date
    Apr 2007
    Location
    Washington State
    Posts
    8,416
    Quote Originally Posted by cypress
    I would think it is appropriate to release just the data and methods required to allow for a completely independent review,analysis and verification of results and conclusions,
    That is not the issue. Insisting that it all be releasable to the general public is the problem. Any independent reviews or analysis and verification need to be done by other scientist anyhow--releasing data and methods to them is MUCH less onerous and follows accepted standards without their own fields of expertise--not the random whims of those who either wouldn't know how to use the data, or who are deliberately trying to immune the process.

    if one wants those conclusions to be seriously considered, particularly when the conclusions can have a large impact on public policy or society norms. Other than that, or if you don't care to have conclusions taken seriously, no need to release data.
    Not a standard for any other kind of science.

    When I think about the practical limitations to such policies I'm astounded. I'll use my own example. I did several technical bulletins and associated papers on the climatology and theory of thunderstorm complexes (mesoscale convective complexes) during the 1980s. An abbreviated version of my Masters Thesis was published in two papers. A combination of the the US Navy, and NOAA paid for most of the work as I was both a graduate student and a civil employee.

    Ok now to the point at hand. At the time, the raw data was available for anyone willing to take the time and write the national data center, NOAA or the GOES satellite coverage. Radar coverage was more difficult but available in NOAA as well. Various other atmospheric scientist reviewed the work both as part of the peer review process to get the papers published or in part as they referenced the work in subsequent peer review papers (about 80 times the last time I checked).

    But lets say two people under the CRU standard tried to dig into that work now?
    The first is another atmospheric scientist, I'll call her Windy. The 2nd is part of the general public who's never published but has a "sincere" interest in the integrity of the system--I'll call him Grumpy.

    Both might start by writing me, the author (or co-author) of the papers. I'd tell them both that all I have left of the original work is the final publications. Windy and Grumpy proceed to ask how to recreate or reveal the data and methods I used.

    Surface and Upper Air data over the Mid West.
    Windy would go back to the sources I mentioned above to recreate the data set.
    Grumpy would insist he gets the actual raw stuff I used. I'd have to tell him the last time I saw it was on a hard drive we used to run on on the university VAX computer from which I did most of my work. The computer is long gone and huge data disk (in physical size), even if available, is almost certainly unreadable even if you found a way to read it.

    Analysis methods.
    I'd tell Windy I used the "Bellamy" method for most of my analysis of the upper air data. Windy would know instant what I meant, because it's a common, efficient and well accepted approach. Grumpy would insist in knowing why I used that particular approach, perhaps claim it as obsolete (from 1949), and if he was dogged and smart enough perhaps points out alternative methods which have better fidelity (in hindsight of course and by computer power hardly available back in the day). Grumpy might ask for the raw code I used and insist I try to find that set of program files on that unreadable disk from the VAX computer and written in Fortran 77--or go through a what might as well be a crate next to the Arc of the Covenant in some warehouse located in Boulder Colorado for the reams of yellowed dot matrix code printouts.

    Than we'd get into radar depictions meticulous traced by a team of low level GS employees from microfilm tapes (8mm I think) are now so brittle it would take an CSI team to recreate. Windy would understand, Grumpy would not.

    Or the satellite imagery of IR cloud top temperatures, also meticulously traced and measured by Planimeter and entered in a data file. Windy would understand, Grumpy would not.

    I could go on and on. The simple fact is raw data and methodologies are best reviewed by other scientist who understand the context, limitations and methods used to conduct the research--as I said before the work I did was reviewed, discussed and conclusions in part refuted, in part accepted and subsumed into the broader understanding of thunderstorm complexes--that's how science is supposed to work. Within the standards of the profession the raw data and methods required were available when it was most relevant. And even now, with a big enough crew and expertise a team of scientist (and budding scientist) could recreate most of the work--but it's not an effort Grumpy could do by himself.

    The CRU standard, which would insist the Grumpies of the world have 100% access to the raw data forever is not only unwise, it's completely unrealistic and prone to gross exploitation of those trying to interfere with the scientific process--or sew seeds of distrust.

    While all scientific communities could probably improve their data methods of archiving and recording meta-data etc, the target audience should always be other scientist like Windy, not the overwhelmingly burdensome process of making everything available to people like Grumpy.
    Meteorologist/Naturalist & Retired Soldier
    “The Holy Land is everywhere” Black Elk
    Reply With Quote  
     

  14. #13  
    Suspended
    Join Date
    Sep 2009
    Location
    New Orleans
    Posts
    1,191
    Quote Originally Posted by Lynx_Fox
    Quote Originally Posted by cypress
    I would think it is appropriate to release just the data and methods required to allow for a completely independent review,analysis and verification of results and conclusions,
    That is not the issue.
    Yes, it is the issue.

    Insisting that it all be releasable to the general public is the problem. Any independent reviews or analysis and verification need to be done by other scientist anyhow--releasing data and methods to them is MUCH less onerous and follows accepted standards without their own fields of expertise--not the random whims of those who either wouldn't know how to use the data, or who are deliberately trying to immune the process.
    Impractical and unworkable for the purpose. Selective release has allowed researchers to withhold data and methods from those capable and willing to look at it most critically. This is why Jones and others are having the difficulty they are having. This is why the public is not buying your arguments. For every crackpot amateur researcher you fear, there are ten more amateur researchers able and willing to show them where they are wrong so that the original research group only has to take on true challenges to their data and methods.

    if one wants those conclusions to be seriously considered, particularly when the conclusions can have a large impact on public policy or society norms. Other than that, or if you don't care to have conclusions taken seriously, no need to release data.
    Not a standard for any other kind of science.

    When I think about the practical limitations to such policies I'm astounded. I'll use my own example. I did several technical bulletins and associated papers on the climatology and theory of thunderstorm complexes (mesoscale convective complexes) during the 1980s. An abbreviated version of my Masters Thesis was published in two papers. A combination of the the US Navy, and NOAA paid for most of the work as I was both a graduate student and a civil employee.

    Ok now to the point at hand. At the time, the raw data was available for anyone willing to take the time and write the national data center, NOAA or the GOES satellite coverage. Radar coverage was more difficult but available in NOAA as well. Various other atmospheric scientist reviewed the work both as part of the peer review process to get the papers published or in part as they referenced the work in subsequent peer review papers (about 80 times the last time I checked).
    Your example is perfect to illustrate the point. at the release of your conclusions, if it was controversial, or significant in some way, some critic with interest would have asked for the data and methods. Now many years later its conclusions are not controversial and so nobody cares that the data and methods are no longer available.

    I could go on and on. The simple fact is raw data and methodologies are best reviewed by other scientist who understand the context, limitations and methods used to conduct the research--as I said before the work I did was reviewed, discussed and conclusions in part refuted, in part accepted and subsumed into the broader understanding of thunderstorm complexes--that's how science is supposed to work. Within the standards of the profession the raw data and methods required were available when it was most relevant. And even now, with a big enough crew and expertise a team of scientist (and budding scientist) could recreate most of the work--but it's not an effort Grumpy could do by himself.
    That's just it, nobody is asking to verify the results because nobody cares enough to do so. If it were important enough, others would also have interest and Grumpy would not be alone in his request. If the data and methods are no longer available then the critics would be justified in arguing the conclusions should be disregarded until they can be verified by either recreating the work or by finding alternate methods.

    The CRU standard, which would insist the Grumpies of the world have 100% access to the raw data forever is not only unwise, it's completely unrealistic and prone to gross exploitation of those trying to interfere with the scientific process--or sew seeds of distrust.
    Nonsense.

    While all scientific communities could probably improve their data methods of archiving and recording meta-data etc, the target audience should always be other scientist like Windy, not the overwhelmingly burdensome process of making everything available to people like Grumpy.
    When cases like the CRU and GISS data sets come up, and the data and methods cannot be verified, then those results should be disregarded or used sparingly until they can be verified or better, alternative verifiable results sets should be used. This is the case of your Grumpy example. Grumpy is unable to obtain the raw data and methods so alternatives are proposed. I note with interest that researchers today are increasingly avoiding use of CRU and GISS results sets in favor of satellite derived results sets and other proxies.

    If the CRU and GISS wishes for their results sets to be given full consideration, then they will have to locate the raw data, re-apply the correction methods, and this time release the information on data and methods so it can be verified. If someone other than those organizations are wishing to help, so much the better. Otherwise the results sets should be viewed with skepticism and should be used with caution.

    Here is an article that seems to indicate that NASA researchers agree their GISS results are derived from data sets that are admittedly in poor shape.
    Reply With Quote  
     

  15. #14  
    Moderator Moderator
    Join Date
    Apr 2007
    Location
    Washington State
    Posts
    8,416
    The reality Cypress as late as two years ago the work was still referenced--long after the realistic or practical recreation of its entire contents was possible.

    The other point is much of the climate data in question, even including the precious satellite information is OLDER than the data I used--the raw data either in completely unusable form or digitized, transcribed, and transformed many times from its original raw form which Grumpy would insist on. Grumpy would claim foul or bend it into conspiracy theory while Windy would understand the verification used in those transformations or simple accept the practical limitations and reality of the situation. Furthermore, only Windy understands the accumulation of potential errors from the initial measuring instrument, calibration requirements, recording of the data into manual form etc etc and most importantly the effect on the final conclusion--while Grumpy without years of education does not.

    PS. This is not a chance to jump into a discussion of GISS data another well beaten horse in these forums.
    Meteorologist/Naturalist & Retired Soldier
    “The Holy Land is everywhere” Black Elk
    Reply With Quote  
     

  16. #15  
    Suspended
    Join Date
    Sep 2009
    Location
    New Orleans
    Posts
    1,191
    Quote Originally Posted by Lynx_Fox
    The reality Cypress as late as two years ago the work was still referenced--long after the realistic or practical recreation of its entire contents was possible.
    And as I previously indicated, we can conclude it is not controversial and therefore nobody wishes to dispute the findings and/or take the time to validate them.

    The other point is much of the climate data in question, even including the precious satellite information is OLDER than the data I used--the raw data either in completely unusable form or digitized, transcribed, and transformed many times from its original raw form which Grumpy would insist on. Grumpy would claim foul or bend it into conspiracy theory while Windy would understand the verification used in those transformations or simple accept the practical limitations and reality of the situation. Furthermore, only Windy understands the accumulation of potential errors from the initial measuring instrument, calibration requirements, recording of the data into manual form etc etc and most importantly the effect on the final conclusion--while Grumpy without years of education does not.
    There is no Grumpy in your case because nobody is disputing or questioning it. If there were actually a Grumpy in your case, without a means to validate the data methods and results, he would be justified in calling for the results to be viewed with caution. The fact that nobody is suggesting this is a reflection of the level of controversy over it (low) and the importance of it in forming controversial policy (also low).

    PS. This is not a chance to jump into a discussion of GISS data another well beaten horse in these forums.
    The CRU issue is an issue of the validity of surface temperature compilations, the reluctance by researchers to release data and methods and the resulting lack of confidence in the results. GISS results are in the same boat though it appears they are on the sinking end of it.
    Reply With Quote  
     

  17. #16  
    Moderator Moderator
    Join Date
    Apr 2007
    Location
    Washington State
    Posts
    8,416
    The House Science and Technology Committee suggest should have access. My point of illustration the practical limitations is that Grumpy would not have the resources or knowledge to get to the truly raw data even it is have been meticulously stored somewhere simply due to conditions of it's initial recorded format. This is true for almost all physical science data more than a couple decades old. The House Science and Technological Committee apparently doesn't understand the practical limitations to it's own recommendation, or is choosing to ignore them. So are you.
    Meteorologist/Naturalist & Retired Soldier
    “The Holy Land is everywhere” Black Elk
    Reply With Quote  
     

  18. #17  
    Suspended
    Join Date
    Sep 2009
    Location
    New Orleans
    Posts
    1,191
    Or more likely, you are incorrect. In the case of CRU and the other surface temperature compilations, these critics who have been after the data and methods are turning out to be correct. These institutions are reluctantly releasing information and as they do, the critics suspicions are being confirmed. This reality argues heavily in favor of full disclosure.
    Reply With Quote  
     

  19. #18  
    Moderator Moderator
    Join Date
    Apr 2007
    Location
    Washington State
    Posts
    8,416
    Quote Originally Posted by cypress
    Or more likely, you are incorrect. In the case of CRU and the other surface temperature compilations, these critics who have been after the data and methods are turning out to be correct. These institutions are reluctantly releasing information and as they do, the critics suspicions are being confirmed. This reality argues heavily in favor of full disclosure.
    Broad based speculation without evidence.

    The simple reality is the raw data is often long gone. What's left has been processed many times usually by methods and a verification process well accepted and fully accessible by the agencies and profession who studies the data. That what needs to be explained to the public--not the creation of unrealistic standards to satisfy the tiniest number of most paranoid conspiracy theorist among the general public that inherently distrust anything others do--not matter how meticulous or detailed.

    Furthermore, even when errors are discovered, that same small number of Grumpy's, usually will exaggerate the effect on what ever conclusions were derived from analysis of the data--heck a good number of the discussions on this forum are about such exaggerations.
    Meteorologist/Naturalist & Retired Soldier
    “The Holy Land is everywhere” Black Elk
    Reply With Quote  
     

  20. #19  
    Suspended
    Join Date
    Sep 2009
    Location
    New Orleans
    Posts
    1,191
    Quote Originally Posted by Lynx_Fox
    Quote Originally Posted by cypress
    Or more likely, you are incorrect. In the case of CRU and the other surface temperature compilations, these critics who have been after the data and methods are turning out to be correct. These institutions are reluctantly releasing information and as they do, the critics suspicions are being confirmed. This reality argues heavily in favor of full disclosure.
    Broad based speculation without evidence.
    The evidence is coming out weekly. The link I provided a bit ago, the one you tried to label as off topic addressed the reality that increasingly those who understand these surface temperature compilations are admitting that they have serious problems.

    The simple reality is the raw data is often long gone. What's left has been processed many times usually by methods and a verification process well accepted and fully accessible by the agencies and profession who studies the data. That what needs to be explained to the public
    Much of the raw data remains, particularly the data that demonstrates these compilations contain serious errors. While those with bias and an agenda attempt to explain the failings away and some may accept the song and dance, the majority has not and I suspect the majority will not.

    --not the creation of unrealistic standards to satisfy the tiniest number of most paranoid conspiracy theorist among the general public that inherently distrust anything others do--not matter how meticulous or detailed.
    Why bring up nonsense like conspiracy theories? I have told you directly that I don't think there is any conspiracy going on. You only give me more opportunity to discredit you when you speak of such nonsense. The majority is sided with me on this, so why speak of the tiniest minority?

    Furthermore, even when errors are discovered, that same small number of Grumpy's, usually will exaggerate the effect on what ever conclusions were derived from analysis of the data--heck a good number of the discussions on this forum are about such exaggerations.
    Human nature is to exaggerate, take yourself and the content of your post as an excellent model of that.

    The reality is that errors have been discovered in these surface temperature result sets, and thus far, there does not appear to be any easy way to correct them in the short term. The wise money is on using alternative compilations until these sets are corrected, and the wise researcher will take a lesson from the mistakes of these people and disclose the situation with their data and methods so they don't loose any credibility they might have.
    Reply With Quote  
     

  21. #20  
    Moderator Moderator
    Join Date
    Apr 2007
    Location
    Washington State
    Posts
    8,416
    What are the noteworthy errors are in the CRU set? Which are so egregious that they change any conclusion?
    Meteorologist/Naturalist & Retired Soldier
    “The Holy Land is everywhere” Black Elk
    Reply With Quote  
     

  22. #21  
    Suspended
    Join Date
    Sep 2009
    Location
    New Orleans
    Posts
    1,191
    Here is a summary of one example of the errors in CRU result sets. It includes a link to an article critical of the result set that was rejected by reviewer Dr. Mann and an email by him to Dr. Jones of the CRU sniggering over the rejection.

    Here is a more technical summary of the problem, again using Siberia in the example.

    Note the several additional links corroborating the errors. Google for more if you wish, there is no shortage of information on this topic.

    It is hard for me to say what if any conclusions you are referring from temperature result sets. I don't see conclusions coming directly from the results so much as what is inferred as a result of using the result sets in other studies. My suggestion here is that the CRU result set should be adjusted to remove this error and the several other errors that are known and now being reported. Until these errors are corrected, alternative result sets should be used. In addition the CRU, and Dr. Jones in particular, have nearly destroyed credibility and they will be hard pressed to restore it.
    Reply With Quote  
     

  23. #22  
    Moderator Moderator
    Join Date
    Apr 2007
    Location
    Washington State
    Posts
    8,416
    Not sure why you are even calling that an error. It's basically a comparison of a new technique to reconcile regional temperature data from dispersed stations (Vincent 1998) compared to the methods CRU uses (HADCRUT2). If you look at the rejected Kamél paper figure 2, it clearly shows most of the difference between the two methods is in the first couple decades of the 20th century. The difference is less than a quarter degree for most of the period in a broad area which shows an increase of 2-4 C by either method.

    It does not establish that there is an error, but suggest a closer look at the particular region, or the urban island assumptions applied to early station data. It does not change any conclusions either. It's quite possible some of the concerns were addressed in the current version (HADCRUT3), though I'm pretty sure the well referenced Vincent method (which I'll look at sometime) has not yet been adopted by the CRU team.

    By the way, if you really are interested, the HADCRU3 data and codes are secretly being concealed from the interested citizens of the world here:
    http://www.cru.uea.ac.uk/cru/data/temperature/
    Meteorologist/Naturalist & Retired Soldier
    “The Holy Land is everywhere” Black Elk
    Reply With Quote  
     

  24. #23  
    Suspended
    Join Date
    Sep 2009
    Location
    New Orleans
    Posts
    1,191
    I call it an error because it is an error. When one fixes an error they generally call the correction a new or improved technique. It's a nice way of saying the old method was in error. The affirmed warming bias is only 0.3-0.5 degrees centigrade so errors ranging between 0-4 C with an mean error of between 0.25 and 0.4 is just about the error one should be looking for.

    I notice that the CRU is still not providing the unadjusted raw data but rather the grid adjusted data.
    Reply With Quote  
     

  25. #24  
    Moderator Moderator
    Join Date
    Apr 2007
    Location
    Washington State
    Posts
    8,416
    Quote Originally Posted by cypress
    I call it an error because it is an error. When one fixes an error they generally call the correction a new or improved technique. It's a nice way of saying the old method was in error. The affirmed warming bias is only 0.3-0.5 degrees centigrade so errors ranging between 0-4 C with an mean error of between 0.25 and 0.4 is just about the error one should be looking for.

    I notice that the CRU is still not providing the unadjusted raw data but rather the grid adjusted data.
    We could get into an entire discussion what you mean by "raw data." But I'm quite sure Grumpy would only be satisfied by the original documents--because you know--perhaps some evil scientist (or clerk) could have changed them. And I'm sure Grumpy would become confused from the massive amount of metadata associated with each station change from moving instruments, replacement boxes along with thermometer calibrations etc. I'm quite sure they'll get busy mailing you temperature record notebooks recorded in pencil on yellowing sheets from the mid 19th century. I wonder how long it would take to comb though that pile of cyrillic records, than make heads of tails of it.

    As they explain on their own web site, much of the raw data is long gone (but copied and digitized after processing to account for the items above) or provided under conditions that it won't be released to others.

    But screw all the practical issues associated with such things--every Tom dick and Harry deserves the raw data and we should spare no expense to replicated it in every detail even after its irretrievable.

    Like most natural sciences the organization that serve as repositories are digitizing much of this kind of data--because it saves time for researchers.

    --
    Different methodologies is not the equivalence as errors. The difference in this case is something on the order of 25% of the observed warming for that tiny area and only applicable if you start in the earliest data set where there's a difference in station selection based on the choice of methodologies. When the the author of the rejected papers doesn't even claims them as errors--how the heck can anyone feel confident enough to make such statements--unless they are mostly interested in exaggerations and cynicism of science instead of objectivity. (Perhaps we should cherry pick a place or two where the different method difference is of the opposite sign--and claim it means the entire set is in "error.")

    It seems, based on actual credible evidence there's really nothing fundamentally wrong with the CRU data set. Naturally, as they have already done, they will continue to improve their product analysis (as well as availability). Given CRU makes their estimated margin of errors available to all, they are very much in keeping with their respective professional standards, as they should be.
    Meteorologist/Naturalist & Retired Soldier
    “The Holy Land is everywhere” Black Elk
    Reply With Quote  
     

  26. #25  
    Forum Isotope Bunbury's Avatar
    Join Date
    Sep 2007
    Location
    Colorado
    Posts
    2,590
    The second report has been published. This is the Oxburgh report of the panel set up by UEA "to assess the integrity of the research published by the Climatic Research Unit in the light of various external assertions".

    It's a very short report. The main conclusions are that there is no evidence of scientific malpractice, and that the CRU probably should have employed professional statisticians rather than try to do the statistical work themselves.

    Perhaps they didn't have sufficient funding to hire professional statisticians.

    http://www.uea.ac.uk/mac/comm/media/...sessment+Panel

    Of course statisticians have had their say on the topic:

    Some organizations that feature these [denialist] views in sophisticated advertising campaigns have manipulated the evidence to create the impression that the consensus among climate scientists is quite different from what it is.
    http://magazine.amstat.org/2010/03/climatemar10/
    Reply With Quote  
     

  27. #26  
    Suspended
    Join Date
    Sep 2009
    Location
    New Orleans
    Posts
    1,191
    Yes it is consistent with what many have been arguing all along. Errors in data handling and processing and improper statistical methods leading to errors including introduction of warming biases. It is not malpractice but the end result is that the temperture results contain introduced errors that should be removed. Statisticians would go a long way in helping to remove introduced biases. Untill the errors are corrected, the results should be used with caution.
    Reply With Quote  
     

  28. #27  
    Forum Isotope Bunbury's Avatar
    Join Date
    Sep 2007
    Location
    Colorado
    Posts
    2,590
    It goes without saying that results should always be treated with caution, as the scientists at CRU went to great lengths to point out.

    The published work also contains many cautions about the limitations of the data and their interpretation.
    Nowhere that I saw in this report were errors mentioned, nor any suggestion that a re-interpretation would negate evidence of warming, so I'm not sure what you mean by that. Neither did the panel find any inappropriate application of statistical method, despite their comment that professional statisticians might have done a better job.

    we did not come across any inappropriate usage although the methods they used may not have been the best for the purpose. It is not clear, however, that better methods would have produced significantly different results.
    It is reasonable for this review panel to mention the issue of statistical method, but for climate denialists who have hounded and vilified the CRU with allegations of dishonesty, fraud and bias, shifting the focus to statistical method amounts to moving the goalposts.
    Reply With Quote  
     

  29. #28  
    Suspended
    Join Date
    Sep 2009
    Location
    New Orleans
    Posts
    1,191
    Quote Originally Posted by Bunbury
    It goes without saying that results should always be treated with caution, as the scientists at CRU went to great lengths to point out.

    The published work also contains many cautions about the limitations of the data and their interpretation.
    Nowhere that I saw in this report were errors mentioned, nor any suggestion that a re-interpretation would negate evidence of warming,
    Not surprising given that was not the focus of the study.

    so I'm not sure what you mean by that.
    Pointing out that critics have noted several mechanisms by which the CRU results contain a warming bias.


    Neither did the panel find any inappropriate application of statistical method, despite their comment that professional statisticians might have done a better job.

    we did not come across any inappropriate usage although the methods they used may not have been the best for the purpose. It is not clear, however, that better methods would have produced significantly different results.
    Right, one generaly does not find what one doesn't search for. Again this was not their purpose. However, they seem to have discovered that statistical controls were not properly applied and without evaluating the impact, sensibly advised better statistical methods.

    It is reasonable for this review panel to mention the issue of statistical method, but for climate denialists who have hounded and vilified the CRU with allegations of dishonesty, fraud and bias, shifting the focus to statistical method amounts to moving the goalposts.
    There have always been a range of opinion from skeptics, some of it close to the mark and some far fetched. The claims of warming biases in the result sets due to improper methods and controls have, from the beginning been one of the more consistent charged against the CRU.
    Reply With Quote  
     

  30. #29  
    Suspended
    Join Date
    Oct 2008
    Posts
    1,847
    Quote Originally Posted by cypress
    The claims of warming biases in the result sets due to improper methods and controls have, from the beginning been one of the more consistent charged against the CRU.
    They haven't been consistent at all, except in their persistence and lack of support in the evidence or careful analysis.

    They have been all over the map, everything from rigging data in various mutually inconsistent ways to manipulating the distribution of source stations, from hiding evidence of cooling to exaggerating evidence of warming, from failing to acknowledge counter evidence from several different sources to inventing supporting evidence of several different kinds to falsely manipulating the evidence they have in vaguely asserted and sensationally described ways.

    The substance of them to date has been political PR, a media campaign conducted through politically friendly blogs and politically oriented marketing efforts in laymen's arena's without backing in rigorous examination of data or statistical breakdown of the criticized "results".
    Reply With Quote  
     

  31. #30  
    Suspended
    Join Date
    Sep 2009
    Location
    New Orleans
    Posts
    1,191
    Wine all you want iceaura, the reports and conclusions all seem to be siding with my descriptions on calculation methods, raw data adjustments and statistical treatment for the CRU and NOAA/GISS results and the Mann Hockey stick graphic.
    Reply With Quote  
     

  32. #31  
    Moderator Moderator
    Join Date
    Apr 2007
    Location
    Washington State
    Posts
    8,416
    Quote Originally Posted by cypress
    Yes it is consistent with what many have been arguing all along. Errors in data handling and processing and improper statistical methods leading to errors including introduction of warming biases.
    Ya that's what the report really said....
    Of course there's no mention of improper handing of data, improper statistical methods, or errors introducing a warming bias--nor any other published credible reports of the same....but no matter: We won't be confused by the facts.

    This tread would be more productive if it actually discussed what the report said, you know things like the unpredictable importance of things and informality of small groups of researchers, why professional statisticians might help ID the most suitable methods (of many suitable but perhaps not best), the value of email and other communications between researches, etc.
    Meteorologist/Naturalist & Retired Soldier
    “The Holy Land is everywhere” Black Elk
    Reply With Quote  
     

  33. #32  
    Suspended
    Join Date
    Sep 2009
    Location
    New Orleans
    Posts
    1,191
    Here is a summary of one aspect of the report corroborating my acount of it. The summary includes a link to the original article.

    The complete report covered several aspects of this issue including use of improper and poor methods and statistical techniques applied to several temperature series including the compilation done by Mann. Note that my post mentioned the Mann temperature series.
    Reply With Quote  
     

  34. #33  
    Forum Isotope Bunbury's Avatar
    Join Date
    Sep 2007
    Location
    Colorado
    Posts
    2,590
    To read what Professor Hand really thinks about this read the FT link. I won't cut and paste here because of copyright and the FT's polite request at the bottom of the article, but please read the paragraphs that start: "Prof Hand said his criticisms should not be seen as invalidating climate science." and then "He accused sceptics of “identifying a few particular issues and blowing them up” to distort the true picture."

    http://www.ft.com/cms/s/0/162b0c58-4...144feab49a.htm

    Hnag on, suddenly it's a subscription required link. But I just read it online for free. Let me see if I can find it again and paraphrase some of it without violating copyright.
    What Prof. Hand actually says is that the handful of errors found so far, including the exaggerated hockey stick graph and a mistaken claim by the IPCC that Himalayan glaciers would disappear by 2035, were “isolated incidents”. And “If you look at any area of science, you would be able to find odd examples like this. It doesn’t detract from the vast bulk of the conclusions."
    Reply With Quote  
     

  35. #34  
    Suspended
    Join Date
    Sep 2009
    Location
    New Orleans
    Posts
    1,191
    The summary I linked to does not come close to attempting to "invalidating climate science". It actually validates climate science done correctly and criticizes poor methods. This is the kind of thing that attempts to keep science honest. It also accurately represents what Dr. Hand actually said and truly thinks about the temperature series. I made no attempt to speculate what Hand thinks about the climate science in general or about those skeptical of AGW except that he did praise those who brought these issues to the forefront.
    Reply With Quote  
     

  36. #35  
    Forum Isotope Bunbury's Avatar
    Join Date
    Sep 2007
    Location
    Colorado
    Posts
    2,590
    And my response was simply to ensure that Hand's remarks are considered in their proper context. His remarks have nothing to do with the report under discussion, in which neither Mann, nor the hockey stick are mentioned. The blog you linked to clearly says Hand is referring to a previous "report" (in fact Mann's original 1998 paper) while you seem to link it to the CRU's independent work. I think it's important to make that clear.
    Reply With Quote  
     

  37. #36  
    Suspended
    Join Date
    Sep 2009
    Location
    New Orleans
    Posts
    1,191
    Oh yes I see that now. I did conflate the two and took them to be connected. Sorry for the error.
    Reply With Quote  
     

  38. #37  
    Veracity Vigilante inow's Avatar
    Join Date
    Oct 2009
    Location
    Austin, TX
    Posts
    3,499
    Quote Originally Posted by cypress
    Oh yes I see that now. I did conflate the two and took them to be connected. Sorry for the error.
    Thank you for this post, cypress.
    Reply With Quote  
     

  39. #38  
    Forum Isotope Bunbury's Avatar
    Join Date
    Sep 2007
    Location
    Colorado
    Posts
    2,590
    Quote Originally Posted by cypress
    Oh yes I see that now. I did conflate the two and took them to be connected. Sorry for the error.
    No problem.
    Reply With Quote  
     

Bookmarks
Bookmarks
Posting Permissions
  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •