Guest article by Douglas Maraun - a replique to von Storch's WSJ op-ed

Recently, my attention has been drawn to the essay "Good Science, Bad Politics" in The Wall Street Journal, written by Hans von Storch. I generally agree with his view on good science, and that action is needed to restore confidence in climate science. But while reading this article, I felt quite concerned - not because I think that all of his criticism is wrong, but rather because some of the points made in the article are not justified. In particular, I do not share the focus on the Climatic Research Unit (CRU) only, where a much more general discussion is necessary. As a former researcher at CRU, I believe some points need some clarification.

First, I want to comment on the term "CRU-cartel" used by von Storch. This term is at least misleading. A cartel is, by definition, a formal agreement. This suggests a rather rigid structure which I believe does not exist. I am not in the position to discuss this issue in depth, but I think, at worst, the term "clique" would be appropriate. "CRU-cartel" furthermore suggests a structure which is centered at CRU, and which comprises all of CRU. Yet there is strong evidence against this point. Von Storch writes about "a concerted effort to emphasize scientific results that are useful to a political agenda by blocking papers in the purportedly independent review process and skewing the assessments of the U.N.'s Intergovernmental Panel on
Climate Change (IPCC). The effort has not been so successful [...]" He does not write that it had been CRU members who actively opposed this effort and ensured that the critique of the hockey stick has been discussed in the IPCC Fourth Assessment Report. Omitting this fact might lead to a biased reception of von Storch's essay.

Second, I agree with von Storch, that some climate scientists are alarmist, and even more, some climate scientists are politicised and give scientific results a certain spin to push their political agenda. Yet, as I experienced CRU, the institute was far from being alarmist or streamlined in any way. In fact, CRU is a small institute of individual characters with very diverse views of the relationship between politics and climate science. And, I want to add, I believe that none of the people working at CRU is alarmist in the sense that they would corrupt their scientific results to push political views.

Third, von Storch writes about the "trick": "we know by now that the activity described by these words was by no means innocent." I disagree with von Storch that the readership of WSJ or the public in general knows what is meant by the trick. In fact, most newspaper articles do not even attempt to explain it, and some even presume that the thermometer measurements had been manipulated (e.g., reports by the german TV
programmes Tagesthemen and Kontrovers imply this prejudice). In fact, the trick was related to a blending of thermometer data into proxy data (reconstructing the temperature of the last millenium), which after around 1960 seem to be dominated by local environmental changes rather than temperature (the so-called divergence problem). Possible problems related to this blending need to be discussed, but von Storch's statement, as vage as it is, might well play into the skeptics claims.

Fourth, the whole transparency issue needs a much broader discussion. For much of the raw data underlying the global average temperature, CRU simply doesn't have the legal right to publish them, even if they wanted to. These data are owned by national weather services and of commercial value and therefore not publicly available. CRU asked all involved national weather services for permission to use these data, and signed not to distribute the raw data further. For details on the data and a further discussion of this point, see
http://www.metoffice.gov.uk/climatechange/science/monitoring/subsets.html. But, and this is quite often overlooked: every critic and skeptic has the possibility to access the data for non-commercial use already now - but they would have to ask all relevant weather services for permission. CRU itself is currently not in the position to publish all of these data.

Finally, the whole discussion - some skeptics claim that what they coined climategate is the biggest scientific scandal in history - should be brought back to earth. It might sound disillusioning, but the formation of cliques is common whereever human beings get organised in a loose way, thus cliques are - unfortunately - common in science. I am a physicist by training, and experienced a clique in a specific subfield of complex systems theory. In that case, some key scientists keep close contact to journal editors, do active gate keeping and suppress critical views. All the issues discussed about the hacked CRU e-mails are almost innocent against the behaviour shown by some scientists in other disciplines, and are far away from real scandals as those about gene therapies in the 1990s, the cloning affair, or the recent manipulation in quantum field theory by a
prospective Nobel candidate. I do not claim that we should accept such circumstances, but we do have to accept that scientists are human beings; and that climate science is - even though it is politically very relevant - not worse than other disciplines which are relevant. I personally believe, and this might sound rather provocative, that climate science is less prone to form cliques and to be corrupted *because* it is politically relevant. Climate science is, as far as I know, the only discipline which is under constant observation by the public and has a transparent and regular assessment of its results (the IPCC reports).

There are, however, some important solely scientific points left, which I believe should be discussed within the climate community.

First, the IPCC has been established by the UNEP and the WMO "to provide the world with a clear scientific view on the current state of climate change and its potential environmental and socio-economic consequences" (IPCC website). In my opinion this also includes reproducibility, and therefore open access to all data flowing into the IPCC reports. If governments want to base their decisions on reliable scientific results, they should allow their national weather services to release all meteorological data used in the IPCC reports, as is already done with the model data. The best solution would be an international data server hosted by the IPCC, which stores and distributes relevant observations. But this is not only a question of working against national egoism and jealousy, but also of money to provide the necessary infrastructure.

Second, I see von Storch's point about communicating to decision makers. Climate scientists have long been working with some kind of "end users". Traditionally these are engineers, which have to rely on climate analyses to build a dam, construct a railway track or plan a new residential area. But more and more, these end users are political decision makers who do not have the time, will and knowledge to understand climate science in detail. This is a true dilemma. On the one hand climate science is complex, and affected by high uncertainties. But on the other hand, the public has the right to understand at least the key points. Not mainly, but also because climate scientists are in general funded by the tax payer. The debate within climate science has long been driven by people who believe that we have to communicate our results as simple as we can such that the general public can understand them. This idea has also been pushed by journalists and university's press offices. During the last years, however, the discussion within the climate community has shifted its focus towards a discussion of uncertainty. This discussion needs to be continued, and we should not allow ourselves to be pushed neither by decision makers nor by journalists.

Third, von Storch claims that "the journals Nature and Science must review their quality-control measures and selection criteria for papers". Nature has a very strict peer review policy (http://www.nature.com/authors/editorial_policies/peer_review.html), and I am not sure whether the system itself can be improved. But still, reviewers and editors should take their roles more seriously within the existing system. I have encountered several editors, who where strongly biased when they had to decide about an article written by a colleague, several reviewers who did not have the time and knowledge to thoroughly review a manuscript, and some reviewers who tried to suppress critical opinions. Editors should abide by their own rules of conduct, and should ensure that also reviewers take their roles seriously. How this can be accomplished is a point to discuss. Nevertheless, I believe that the overall system works fairly well. On the one hand, claims that certain results cannot be published because of a conspiracy are pathetic; a good paper rejected by one journal will always have the chance to be published by another one. On the other hand those hundreds of bad papers published every year do not hamper the scientific discussion; one paper is only one single contribution to a discussion, not a complete and accepted theory.

Print this post