Over the last couple of months the IPCC’s Fourth Assessment Report (published in February 2007) has been at the centre of a media storm. Revaluations about exaggerated or groundless claims have called into question the reputation of an organisation that has assumed a mantle of scientific invincibility during the last three years.
Alarmist predictions about the future of Himalayan glaciers, the Amazon rain forests, agricultural production in Africa, increasing devastation caused extreme weather events and rising sea levels have been shown to be based on evidence that at best is anything but robust and at worst is no more than hearsay. Worse still, it seems that the authors of the report were aware of the shortcomings of the evidence they were relying on but used it anyway.
Publication on the Internet of over a thousand emails from the Climatic Research Unit at the University of East Anglia, now known as Climategate, has added to disquiet about the IPCC’s activities. They suggest that Professor Phil Jones and many other leading climate scientists have attempted to subvert the accepted standards of their profession in order to protect their research findings from criticism. Many of those involved have been extremely influential within the IPCC process and the emails reveal an unhealthy culture of hostility towards anyone who questions the orthodox view of climate change that this organisation represents. It is questionable whether objective scientific research can take place under such circumstances.
The effect on the IPCC’s reputation, and that of its chairman Dr Rajendra Pachauri, has been devastating, but at every stage of this scandal we have been assured that the core science underpinning concern about anthropogenic climate change has remained unscathed. The IPCC and its supporters have been able to undertake this damage limitation exercise because attention so far has focused on only one of the three sections of the most recent assessment report: Working Group II: Impacts, Adaptation and Vulnerability. This deals with the symptoms and perceived consequences of climate change. The core scientific evidence that the climate is changing and that human influence is playing a part in this is contained in another section of the report, Working Group I: Climate Change 2007: the Physical Basis. But can we be confident that the same problems of sloppy authorship and exaggeration do not extend to this part of the IPCC’s assessment too?
On page 8 of the Working Group I: Summary for Policymakers there is a table (SPM.2) that has the following snappy title:
Recent trends, assessment of human influence on the trend and projections for extreme weather events for which there is an observed late-20th century trend.
Extreme weather events are a particularly potent weapon in the battle to win hearts and minds for the crusade against climate change. Hurricanes (cyclones), droughts, floods and heat waves all feature as dramatic news stories regularly; they are the stuff of which editor’s dreams are made. A combination of dramatic pictures and human interest assures them a place in the headlines, and on the front pages, whenever there is an excuse for publishing them. From the point of view of those who wish to promote the idea of anthropogenic global warming such stories present a wonderful opportunity. If extreme weather events can be linked to catastrophic climate change in the public consciousness then the message that humans are influencing the climate, with appalling consequences, is going to be reinforced repeatedly because hardly a month goes by without some kind of weather related disaster being reported.
This is what Table SPM.2 says:
Phenomenon aand direction of trend
Likelihood that trend occurred in late 20th century (typically post 1960)
Likelihood of a human contribution to observed trend b
Likelihood of future trends based on projections for 21st century using SRES scenarios
 Warmer and fewer cold daysand nights over most landareas
Very likely c
Virtually certain e
 Warmer and more frequenthot days and nights overmost land areas
Very likely d
Likely (nights) e
Virtually certain e
 Warm spells / heat waves.Frequency increases overmost land areas
More likely than not f
 Heavy precipitation events.Frequency (or proportion oftotal rainfall from heavy falls)increases over most areas
More likely than not f
 Area affected by droughtsincreases
Likely in many regions
More likely than not
 Intense tropical cycloneactivity increases
Likely in some regions
More likely than not f
 Increased incidence ofextreme high sea level(excludes tsunamis) g
More likely than not f, h
For ease of reference I’ve made a few additions to this table in square brackets. The columns are now labelled A-D and the rows 1-7. I’ve also added numerical values for probabilities (‘likelihood’). These are explained briefly in a footnote on page 3 of the Summary for Policymakers, and in more detail in Box TS.1 in the Technical Summary of Working Group I as follows:
Likelihood Terminology Likelihood of the occurrence/ outcome Virtually certain > 99% probability Extremely likely > 95% probability Very likely > 90% probability Likely > 66% probability More likely than not > 50% probability About as likely as not 33 to 66% probability Unlikely < 33% probability Very unlikely < 10% probability Extremely unlikely < 5% probability Exceptionally unlikely < 1% probability
If we look at row  in Table SPM.2 for example, which deals with heat waves, we find the following:
Column A describes a phenomenon and a trend: heatwaves becoming more frequent.
Column B assesses the ‘likelihood’ that this trend has been confirmed by observation: is there empirical scientific evidence that heatwaves have become more frequent?
Column C introduces a hypothesis: if a trend in the frequency of heat waves has been observed then human activity is contributing to that trend. The likelihood of this being true is assessed.
Column D considers a prediction about the trend – increasing frequency of heat waves – continuing during the rest of the 21st century. This prediction is based on a range of scenarios set out in the SPM from Page 18 onwards. All of these envisage a world in which the fraction of Co2 in the atmosphere is growing as a result of human activity with a consequent rise in global temperatures exceeding any natural variation that can be expected.
In terms of the scientific method, we start with an assumption in Column [A]; heatwaves are increasing. This is then tested by observation in Column [B]: is it possible to detect an increase in the number of heatwaves, particularly during the last fifty years? A hypothesis follows in Column [C]: if it can be shown that there has been an increase in the number of heat waves, then anthropogenic warming has contributed to this. Finally, in Column [D], there is a prediction that depends on the preceding columns.
If we now look at the levels of ‘likelihood’ that the IPCC have assigned in each column we find something really quite remarkable. There is only a 60% – 89% chance that the frequency of heat waves have in fact increased; this leaves significant room for doubt that any such trend exists. There is even less confidence (50% – 59%) that, if an increase in heat waves has occurred – and we’re not sure it has – then human activity has something to do with it.
The conclusion that the IPCC draws from this is that, although there is a significant level of uncertainty as to whether the frequency of heat waves has increased during the last half century, and there is even more uncertainty as to whether, if the frequency has in fact increased, this can be attributed to human influence, a prediction can be made that heatwaves will increase during the next ninety years as a result of anthropogenic global warming. The ‘likelihood’ assigned to this is of 90-94%. Therefor according to the IPCC, confidence in the prediction is higher than confidence in either the observations or the hypothesis that the prediction is based on.
This makes no sense to me, but then I am not a scientist, let alone a climate scientist. It would be very interesting to hear the views of researchers from other disciplines, not on the merits of the scientific evidence, but as to whether this table does in fact defy logic.
Scanning the levels of ‘likelihood’ expressed in the rest of the table shows a remarkable consistency. The ‘likelihood’ expressed in Column [B] is higher than that in Column [C], but Column [D] is higher than in Column [C]. In some instances the assessment of ‘likelihood’ in Column [D] is higher than in either Columns [B] or [C] . So in each case the IPCC are saying that they have greater confidence in the predictions than in the hypothesis on which it is based, and in some cases that confidence in the prediction is even higher than in the observed trend.
The Working Group I Summary for Policymakers is intended to present a transparent and objective assessment of whether anthropogenic global warming is taking place, and do so in a way that is accessible to laypeople who would be unable to draw their own conclusions from the thousands of pages of scientific references in the main report. To a great extent this requires that policy makers must trust the scientists who write the IPCC’s reports, and they are likely to do so. We are repeatedly assured that the IPCC’s findings are based on the most carefully reviewed scientific research and therefore there is no room for argument about the facts. This point of view is epitomised by the oft-repeated mantras, ‘the science is settled’ and ‘you can’t argue with the science’.
The take-away message for anyone reading Table SPM.2 is that the IPCC attaches a high level of confidence to predictions that anthropogenic warming will cause extreme weather events to increase during the current century. They do not need to add that droughts, floods, hurricanes and heat waves cause loss of life and wreak economic havoc and misery on those who experience them. Regular news converge of such events leaves us in no doubt about this. Such events are the most visible and dramatic symptoms of our restless climate and we are well aware of their consequences.
The IPCC’s message to policy makers is clear: reduce greenhouse gas emissions now or more people will get hurt, and it will be your fault. So it is very important that those who are in a position to allocate billions of pounds, dollars or euros to fighting global warming should be fully aware of the methodology that the IPCC is using to reach this conclusion, and particularly just what confidence can be placed in their predictions.
The footnote on Page 3 of the Working Group I Summary for Policymakers that describes how confidence in scientific understanding and predicted outcomes is assessed says this:
In this Summary for Policymakers, the following terms have been used to indicate the assessed likelihood, using expert judgement, of an outcome or a result …
Therefore the predictions of ‘likelihood’ in Table SPM.2 are not derived from any specific research contained in the assessment report but rely on the views of the authors of the report. So we are not talking about confidence levels that have been arrived at mathematically here, but the opinions of scientists. And reliance on their opinions is far reaching.
A minuscule footnote (f) to Table SPM.1 says:
Magnitude of anthropogenic contributions not assessed. Attribution for these phenomena based on expert judgement rather than formal attribution studies.
This refers to the predictions in Column [C] rows 3,4,6 and 7. It would appear that so far as these categories of extreme weather events are concerned there are no studies cited in the report that attribute human influence to the phenomena concerned, and the claims made in the table are based entirely on the ‘expert judgement’ of the authors.
Looking at the notes accompanying Table SPM.2 we find that it summarises the findings of Chapter 3 of the Working Group I section of the report: Observations: Surface and Atmospheric Climate Change . The coordinating lead authors of this chapter were Dr Kevin E Trenberth of National Center for Atmospheric Research (NCAR), and Professor Phil Jones, at that time director of he Climatic Research Unit (CRU) at the University of East Anglia.
The CRU emails leaked on the internet before Christmas made Professor Jones the unwilling star of what has been described as the biggest scientific scandal in living memory. He has now stepped down as director of the CRU pending inquiries into Climategate.
Dr Trenberth was also one of the correspondents featured in the Climategate emails and he has this to say (12/10/2009) in a message to Professor Michael Mann of Hockey Stick graph fame concerning the present decade long standstill in global warming;
Well I have my own article on where the heck is global warming? We are asking that here in Boulder … [Colorado, home to NCAR]
The fact is that we can’t account for the lack of warming at the moment and it is a
travesty that we can’t. The CERES data published in the August BAMS 09 supplement on 2008 shows there should be even more warming: but the data are surely wrong. Our observing system is inadequate.
This suggests that Dr Trenberth’ reaction to observed data that fail to match predictions is to reconsider the observations rather than question the skill of the predictions; something that I think most research scientists would find alarming.
(This message was copied to Stephen H Schneider , Myles Allen , Peter Stott , Phil Jones , Benjamin Santer , Tom Wigley , Thomas R Karl , Gavin Schmidt , James Hansen and Michael Oppenheimer. Although it is outside the scope of this post, it is worth noting that three of the recipients are involved in compiling global surface temperature records: Jones (CRU), Karl (NOAA), and Hansen (NASA GISS)).
In 2005 Dr Trenberth organised a major ‘media event’ at which he informed the world’s press that global warming was causing increased hurricane activity. This led to the resignation from the IPCC of hurricane expert Chris Landsea who was the contributing author responsible for the hurricanes section of Working Group I Chapter 3 of the Fourth Assessment Report. Prior to the press conference, Landsea warned Trenberth - who was not a hurricane expert - that there were no credible research findings that supported this view, but his warning was ignored. When Landsea complained to the governing council of the IPCC that making claims without sound scientific evidence would prejudice the organisation’s credibility their response was to defend Trenberth’s behaviour.
The extent of the scandal revealed by the CRU emails is beyond the scope of this post, but one quotation from a message written by Professor Jones would seem to be particularly relevant so far as the matter of ‘expert judgement’ is concerned:
As you know, I’m not political. If anything, I would like to see the climate change happen, so the science could be proved right, regardless of the consequences. This isn’t being political, it is being selfish.
Does this suggest a frame of mind in which one can reasonably expect a scientist to exercise his ‘expert judgement’ objectively? There is plenty of circumstantial evidence in the CRU emails suggesting that Jones may have behaved improperly both in connection with his own research and his contributions to the IPCC. Climategate, and Jones’ conduct, are now the subject of inquiries by both the University of East Anglia and The House of Commons Select Committee on Science and Technology.
As well as being the Coordinating Lead Authors on Working Group I Chapter 3, both Trenberth and Jones were Draft Contributing Authors for the Working Group 1 Summary for Policymakers. It is inconceivable that they would not have been involved in the creation and inclusion of Table SPM.2 in that summary.
One of the most worrying aspects of the flaws in the IPCC’s Working Group II report, which have already received so much publicity, is that they have only come to light because they were publicised by bloggers. As I said earlier in this post, I am not a scientist and there may be perfectly reasonable explanations for the apparent inconsistencies that Table SPM.2 reveals that I have missed. So far as I am aware there is no one involved with the IPCC who I can ask about this in the expectation of getting a fair and objective response. If the Working Group I report is to be scrutinised, then this too is only likely to happen if bloggers persistently ask awkward questions.
At the beginning of this week Professor Jones gave evidence before the House of Commons Science and Technology Committee. Andrew Orlowski of The Register ended a very perceptive report of the proceedings by concluding that the committee may have come to the conclusion ‘that rotten scientists perhaps mean rotten science’. At the time that the IPCC’s Fourth Assessment Report was published there was no one outside the sceptical bogosphere who was likely to entertain the idea that this vast and heavily hyped document might contain egregious errors. All that has changed now with the revelations concerning Working Group II. It is high time that the same detailed analysis is applied to the findings of Working Group I, the part of the report that is supposed to provide the real evidence for anthropogenic global warming. Furthermore, where the behaviour of scientists as revealed by Climategate is in doubt, their input to the IPCC reports should be scrutinised, otherwise the IPCC’s claims that none of the revelations so far have weakened the essential evidence of anthropogenic global warming will be credible. The address headers of the CRU emails read like a list of the IPCC hierarchy.
Here is one last word about the IPCC use of ‘expert judgement’ in assessing the probability of scientific understanding and likelihood’s. The United Nations Environment Programme is one of the parent bodies of the IPCC, with the World Meteorological Organisation (WMO). UNEP handled the publicity for the February 2007 press release of the Working Group I Summary for Polichymakers. This is the headline that they used:
Evidence of Human-caused Global Warming “Unequivocal”, says IPCC
The media worldwide dutifully ran headlines and stories that said exactly that, but the term ‘unequivocal’ only appears once in the Summary for Policymakers in the following context:
Warming of the climate system is unequivocal …
The supposed human contribution to this warming, which amount to significantly less than 1oC is assessed as ‘very likely’, which falls a long way short of ‘unequivocal’, so it appears that UNEP were prepared to ‘sex up’ this report from the outset.
In the IPCC’s Third Assesment Report, published in 2001, the probability that humans are contributing to climate change was assessed as ‘likely’. It would be very interesting to know whose ‘expert judgement’ led to the probability being increased to ‘very likely’ in 2007, and on what considerations this decision was based.