House of Commons - The disclosure of climate data from the Climatic Research Unit at the University of East Anglia

The disclosure of climate data from the Climatic Research Unit at the University of East Anglia - Science and Technology Committee Contents

2 Datasets

Climate science

18. Climate is distinct from weather: it is the average of weather conditions over a number of years. Climatologists study climates in different parts of the world and for the Earth as a whole. CRU, according to its website: "has developed a number of the data sets widely used in climate research, including the global temperature record used to monitor the state of the climate system, as well as statistical software packages and climate models".[31]

19. The process of calculating the Earth's average global temperatures (past, present and future) is complicated and lengthy. Data from thousands of weather stations all around the world, on land and at sea, must be collected, checked for quality, adjusted for inconsistencies and error margins, and then mapped onto a series of grids on the Earth's surface. The methods, results and conclusions are then presented to the academic world, first by passing the peer review process prior to publication, and second, after presentation, the scrutiny of the wider academic community.

20. Climate science, like any other science, uses the scientific method to make its assessments of past and present climate and predictions about the future climate. The key characteristics of the scientific method can be described as: characterisations, hypotheses, predictions, and experiments.

· Characterisations: consideration of a problem, and examination of whether or not an explanation exists for it.

· Hypotheses: if no such explanation exists, a new explanation is stated.

· Predictions: what consequences follow from a new explanation?

· Experiments: is the outcome consistent with the predicted consequences?

Each of these is subject to peer review prior to the formal sharing of knowledge through publication. Through peer review scientists allow their views and methods to be critically appraised expertly and externally.

· Replication and verification

To have the results and conclusions survive criticism or scepticism and be part of the accepted canon of scientific knowledge, most experiments will have to be demonstrably replicable (by the same group) to pass peer review and will often need to be verified by other independent researchers taking similar approaches.

21. Therefore climatologists are, like other scientists, required to test their theories—such as global warming and the causes of warming—against observational data. They must also replicate and verify their experiments, by holding independent datasets and conducting independent analyses of these datasets, and by publishing their full methods and results for scrutiny. Ultimately, these ideas are put up to the threat of falsification by other scientists working in the field.

22. In this Chapter we discuss some aspects of this process.

Context

23. There are three main international climate datasets, which have been built up from direct temperature measurements on land and sea at weather stations all around the world:

a) the National Climatic Data Center (NCDC) of the National Oceanographic and Atmospheric Administration (NOAA) in Asheville, North Carolina, USA;

b) the Goddard Institute of Space Studies (GISS), part of the National Aeronautic and Space Administration (NASA) in New York, USA; and

c) CRUTEM3, at CRU, UEA.[32]

24. In addition, there are two others, one in Russia and one in Japan, that use similar methods.[33] There are also two that use satellite observations, by the University of Alabama at Huntsville and by Remote Sensing Systems, California.[34]

25. Professor Jones, commenting on the different climate research groups around the world in the UK, US, Russia and Japan,[35] told us that:

26. What sets the CRU dataset apart is its comprehensiveness:

Complaints and accusations

27. The complaints and accusations made against CRU in relation to the scientific process come under two broad headings. The first is transparency: that CRU failed to abide by best scientific practice by refusing to share its raw data and detailed methods. The second is honesty: that CRU has deliberately misrepresented the data, in order to produce results that fit its preconceived views about the anthropogenic warming of the climate. We take each of these complaints and accusations in turn.

TRANSPARENCY

Raw data

28. Warwick Hughes, a "freelance earth scientist from Australia",[38] had asked Professor Jones for CRU's raw data. He received the following reply:

29. On the face of it, this looks like an unreasonable response to a reasonable request. As Lord Lawson put it: "Ask any decent scientist and they will say the keystone for integrity in scientific research is full and transparent disclosure of data and methods".[40] However, Professor Jones, while confessing that he has sent some "awful" e-mails,[41] defended his position.

30. First, in answer to the question of whether the raw data are accessible and verifiable, Professor Jones told us that:

31. In addition, of course, there are the sources of the data, the weather stations, to which any individual is free to go and collect the data in the same way that CRU did. This is feasible because the list of stations that CRU used was published in 2008.[43]

32. Even if CRU had wanted to, it would have been unable to publish all of these data because, as Professor Acton explained, some of the data are bound by commercial agreements with different national meteorological organisations:

(The issue with Sweden has since been resolved. The Swedish Meteorological and Hydrological Institute gave permission for CRU to publish its Swedish data on the UEA website on 8 March 2010.[45])

33. Second, as UEA explained in its submission, it is:

34. Professor Jones added, when he gave oral evidence:

35. These kinds of adjustments to raw data take a lot of time. That is why, in the words of Professor Jones, "Most scientists do not want to deal with the raw station data, they would rather deal with a derived product".[48]

36. A third point was made by Professor Acton that CRU should not be under any obligation to provide raw data:

37. CRU's refusal to release the raw data gave some the impression that it was deliberately keeping its work private so that its studies could not "be replicated and critiqued".[50] The Peabody Energy Company said of CRU that "they appeared to be particularly concerned that putting their information in the public domain would expose their work to criticism".[51] Even an effort to conduct a simple quality check was said to be thwarted by CRU's unwillingness to share the data it had used.[52] In contrast, NASA has been able to make all its raw data available as well as its programmes.[53]

38. We recognise that some of the e-mails suggest a blunt refusal to share data, even unrestricted data, with others. We acknowledge that Professor Jones must have found it frustrating to handle requests for data that he knew—or perceived—were motivated by a desire simply to seek to undermine his work. But Professor Jones's failure to handle helpfully requests for data in a field as important and controversial as climate science was bound to be viewed with suspicion. He was obviously frustrated by other workers in the field trying to "undermine" his work, but his actions were inevitably counterproductive. Professor Jones told us that the published e-mails represented only "one tenth of 1%" of his output, which amounts to one million e-mails, and that we were only seeing the end of a protracted series of e-mail exchanges. We consider that further suspicion could have been allayed by releasing all the e-mails. In addition, we consider that had the available raw data been available online from an early stage, these kinds of unfortunate e-mail exchanges would not have occurred. In our view, CRU should have been more open with its raw data and followed the more open approach of NASA to making data available.

39. We are not in a position to set out any further the extent, if any, to which CRU should have made the data available in the interests of transparency, and we hope that the Independent Climate Change Email Review will reach specific conclusions on this point. However, transparency and accountability are of are increasing importance to the public, so we recommend that the Government reviews the rules for the accessibility of data sets collected and analysed with UK public money.

Methods

40. The Royal Society of Chemistry in its submission made it clear that:

There have been criticisms that Professor Jones and colleagues have not shared their methodologies. Andrew Montford, author of The Hockey Stick Illusion,[55] pointed out in his memorandum that:

41. Professor Jones contested these claims. According to him, "The methods are published in the scientific papers; they are relatively simple and there is nothing that is rocket science in them".[57] He also noted: "We have made all the adjustments we have made to the data available in these reports[58]; they are 25 years old now".[59] He added that the programme that produced the global temperature average had been available from the Met Office since December 2009.[60]

42. On this basis, he argued, it was unnecessary to provide the exact codes that he used to produce the CRUTEM3 chart. The Met Office had released its code and it produced exactly the same result.[61]

43. In answer to the charge that the computer codes that were stolen from CRU's computer network were defective,[62] Professor Jones pointed out that:

44. CRU's alleged refusal to disclose its assumptions and methodologies gave credence to the view that exposure to "independent scrutiny would have undermined the AGW [anthropogenic global warming] hypothesis".[64] However, the failure to publish the computer code for CRUTEM3 left CRU vulnerable when concerns emerged that other codes it used had faults. John Graham-Cumming, a professional computer programmer, told us that:

45. The conspiracy claims were fuelled by CRU's refusal to share the most detailed aspects of its methodologies, for example, the computer codes for producing global temperature averages. We note that the research passed the peer review process of some highly reputable journals. However, we note that CRU could have been more open at that time in providing the detailed methodological working on its website. We recommend that all publicly funded research groups consider whether they are being as open as they can be, and ought to be, with the details of their methodologies.

Repeatability and verification

46. These complaints and concerns surrounding transparency cut to the heart of the scientific process. It has been argued that without access to the raw data and detailed methodology it is not possible to check the results of CRU's work. The Institute of Physics pointed out that:

47. This has substance if one considers CRU's work in isolation. But science is more than individual researchers or research groups. One should put research in context and ask the question: what would one hope to find by double checking the processing of the raw data? If this were the only dataset in existence, and Professor Jones's team had been the only team in the world to analyse it, then it might make sense to double check independently the processing of the raw data and the methods. But there are other datasets and other analyses that have been carried out as Professor Jones explained:

48. In its memorandum UEA explained the differences between the methodologies used by three basic datasets for land areas of the world, NOAA, NASA and CRU/UEA:

49. In our call for evidence we asked for submissions on the question of how independent the other international data sets are. We have established to the extent that a limited inquiry of this nature can, that the NCDC/NOAA and GISS/NASA data sets measuring temperature changes on land and at sea have arrived at similar conclusions using similar data to that used by CRU, but using independently devised methodologies. We have further identified that there are two other data sets (University of Alabama and Remote Sensing Systems), using satellite observations that use entirely different data than that used by CRU. These also confirm the findings of the CRU work. We therefore conclude that there is independent verification, through the use of other methodologies and other sources of data, of the results and conclusions of the Climate Research Unit at the University of East Anglia.

50. The fact that all the datasets show broadly the same sort of course of instrumental temperature change since the nineteenth century compared to today was why Professor John Beddington, the Government Chief Scientific Adviser, had the confidence to say that human induced global warming was, in terms of the evidence to support that hypothesis, "unchallengeable":[74]

51. 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.

52. That is probably part of why it has not been practice in the climate science community to publish all the data and computer codes with the academic papers. We got to the crux of the issue during an interesting exchange with Professor Jones:

Graham Stringer:

Professor Jones:

Graham Stringer:

Professor Jones:

Graham Stringer:

Professor Jones:

53. Another reason why data and the codes were not published may be that norms for publication evolved in a period when the journals were only published in hard copy. In such circumstances it is understandable why an editor would not want to publish raw climate data (extremely long lists of numbers) and code for the computer programmes that analyse the data (which run to hundreds of thousands of lines of code). However, in the age of the internet, these kinds of products can be made available more easily, and we are minded to agree with Professor Jones observation on this point that: "Maybe it should be standard practice".[77]

54. It is not standard practice in climate science and many other fields to publish the raw data and the computer code in academic papers. We think that this is problematic because climate science is a matter of global importance and of public interest, and therefore the quality and transparency of the science should be irreproachable. We therefore consider that climate scientists should take steps to make available all the data used to generate their published work, including raw data; and it should also be made clear and referenced where data has been used but, because of commercial or national security reasons is not available. Scientists are also, under Freedom of Information laws and under the rules of normal scientific conduct, entitled to withhold data which is due to be published under the peer-review process.[78] In addition, scientists should take steps to make available in full their methodological workings, including the computer codes. Data and methodological workings should be provided via the internet. There should be enough information published to allow verification.

DISHONESTY

55. Of all the e-mails released, one dated 16 November 1999 has caused particular concern:

sic

56. The word "trick" and the phrase "hide the decline" have been taken by some to demonstrate intent on the part of Professor Jones to "falsify data" and to "exaggerate warming".[80]

"Trick"

57. In his submission, Peter Taylor, author of Chill,[81] states that:

58. UEA interpreted the use of the word "trick" differently:

59. These interpretations of the colloquial meaning of "trick" have been accepted by even the staunchest of critics:

Lord Lawson of Blaby:

Chairman:

Lord Lawson of Blaby:

60. 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.

"Hide the decline"

61. Lord Lawson did, however, describe CRU's treatment of the data as "reprehensible",[85] because, in his view, Professor Jones deliberately hid data that demonstrated a decline in temperatures.[86]

62. The data that he believed to be "hidden" are a set of tree ring data that disagree with other data sources regarding temperature trends. Lord Lawson said: "when the proxy series [...] departed from the measured temperature series, a normal person will say maybe that means the proxy series is not all that reliable".[87] In that context he made two specific claims:

· that the tree ring data were flawed because "for a long period before 1421 they relied on one single pine tree";[88] and

· that the divergence problem was not just for data after the 1960s, "it is not a good fit in the latter half of the nineteenth century either".[89]

63. It is outside the remit of the terms of reference of this inquiry to make a detailed assessment of the science, but it is worth noting that Professor Jones had a very different perspective. On the first point, he commented:

64. On the second point, he told us:

65. Professor Jones has published on this issue on several occasions, including a 1998 Nature paper[92] and subsequent papers.[93] He contested the view that he was trying to hide the decline in the sense that he was trying to pretend that these data did not exist and thereby exaggerate global warming: "We do not accept it was hidden because it was discussed in a paper^[[94]^] the year before and we have discussed it in every paper we have written on tree rings and climate".[95] Rather, what was meant by "hide the decline" was remove the effects of data known to be problematic in the sense that the data were known to be misleading. UEA made it clear in its written submission that:

66. 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.

PERVERTING THE PEER REVIEW PROCESS

67. The main allegations on the suppression or distortion of others' findings concern the role of CRU in the operation of the peer review process. It has been alleged that scientists at CRU abused the peer review process to prevent those with dissenting views on climate change the opportunity in getting papers published. There are three key accusations. First, David Holland, an author of several FOIA requests that were mentioned in the leaked e-mails, claimed that climate scientists at CRU corrupted the IPCC process:

68. In one e-mail, Professor Jones appeared to suggest that he and another scientist would deliberately try to "keep out" two papers from the IPCC's Fourth Assessment Report.[98]

69. The second is that climate scientists tried to suppress a paper on research fraud. As Dr Benny Peiser, Director of the Global Warming Policy Foundation, put it:

70. The third allegation is made by Dr Sonja Boehmer-Christiansen, a former peer reviewer for the IPCC, editor of the journal, Energy & Environment, and Reader Emeritus at Hull University, who stated in her memorandum that her journal became the focus of attacks from CRU scientists:

100

71. When we asked Professor Jones about these accusations, he contested each of them.

· On the claim that he tried to keep two papers out of the IPCC report, he explained that the papers were already published and that "I was just commenting that I did not think those papers were very good".[101]

· On the claim by he tried to suppress papers that alleged research fraud, he told us:

Energy & Environment

102

· On the claims made by Dr Boehmer-Christiansen, he noted: "I was sending an email to the head of department about a complaint that she had made about me to the UK Climate Impacts Programme, so I was just responding there".[103]

72. In summary, Professor Jones argued:

104

73. 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.

31 www.cru.uea.ac.uk/cru/about Back

32 Ev 21, para 4.2 Back

33 Q 78 Back

34 Ev 104 [D.R. Keiller], para 2 Back

35 Q 79 Back

36 Q 80 Back

37 Ev 64 [John Beddington and Julia Slingo] Back

38 www.warwickhughes.com Back

39 Ev 158, Appendix 1 Back

40 Q 9 Back

41 Q 103 Back

42 Q 78 Back