Sunday, February 07, 2010

And now for Africagate

Following an investigation by this blog (and with the story also told in The Sunday Times), another major "mistake" in the IPCC's benchmark Fourth Assessment Report has emerged.

Similar in effect to the erroneous "2035" claim – the year the IPCC claimed that Himalayan glaciers were going to melt – in this instance we find that the IPCC has wrongly claimed that in some African countries, yields from rain-fed agriculture could be reduced by up to 50 percent by 2020.

At best, this is a wild exaggeration, unsupported by any scientific research, referenced only to a report produced by a Canadian advocacy group, written by an obscure Moroccan academic who specialises in carbon trading, citing references which do not support his claims.

Unlike the glacier claim, which was confined to a section of the technical Working Group II report, this "50 percent by 2020" claim forms part of the key Synthesis Report, the production of which was the personal responsibility of the chair of the IPCC, Dr R K Pachauri. It has been repeated by him in many public fora. He, therefore, bears a personal responsibility for the error.

In this lengthy post, we examine the nature and background of this latest debacle, which is now under investigation by IPCC scientists and officials.

The tale unfolds

"Excellencies, members of the media, distinguished ladies and gentlemen! I speak to you in the voice of the world's scientific community...". So declared Dr R K Pachauri in his opening address to the climate summit in Potsdam last September.

In the name of that "scientific community" which in November 2007 had completed the IPCC's Fourth Assessment Report (AR4) - "the collective effort of almost four thousand of the world's best specialists working tirelessly over five years" – Dr Pachauri larded his speech with examples of impending doom.

Thus he told the assembly that, by 2020, "in some countries of Africa yields from rain-fed agriculture could be reduced by up to 50 percent".

Crucially, this was not a random statement plucked from one of the working group reports but one highlighted in the "gold standard" Synthesis Report (Section 3.3.2).

That report is based on the assessment carried out by the three Working Groups and provides "an integrated view of climate change" as the final part of the IPCC's Fourth Assessment Report (AR4). It represents the considered view of the IPCC as a corporate body and is the part of AR4 which Pachauri had personally supervised as leader of the core writing team.

The claim, as set out in that key report, is apocryphal in tone and content. Applying to Africa as a whole, it tells us:
By 2020, in some countries, yields from rain-fed agriculture could be reduced by up to 50%. Agricultural production, including access to food, in many African countries is projected to be severely compromised. This would further adversely affect food security and exacerbate malnutrition.
The immediate source of this claim is the Working Group II report, Chapter 9 (Pg 448) on Africa, where we see the following:
In other countries, additional risks that could be exacerbated by climate change include greater erosion, deficiencies in yields from rain-fed agriculture of up to 50% during the 2000-2020 period, and reductions in crop growth period (Agoumi, 2003).
This is actually a more nuanced claim than is found in the Synthesis Report, but it is here that problems begin to be apparent, specifically in the "Agoumi" reference. It goes to this document:
Agoumi, A., 2003: Vulnerability of North African countries to climatic changes: adaptation and implementation strategies for climatic change. Developing Perspectives on Climate Change: Issues and Analysis from Developing Countries and Countries with Economies in Transition. IISD/Climate Change Knowledge Network, 14 pp.
Advocacy group origins

The IISD is the International Institute for Sustainable Development. Although it calls itself a "think tank", it is another environmental advocacy group in the same mould as the WWF.

Established in Canada in 1990, it was the brainchild of then prime minister Brian Mulroney, intended to be "a centre which will promote internationally the concept of environmentally sustainable development." As this hagiography makes abundantly clear, its primary concern is policy advocacy.

It is unashamedly strident in declaring its "vision" to be: "Better living for all - sustainably". Its "mission", set out in evangelistic tones, is: "To champion innovation, enabling societies to live sustainably." The website tells us:
The world is challenged by a changing climate, biodiversity loss, abject poverty and environmental degradation. What can make a difference? Good ideas. Creativity. Passion. Innovation. The achievement of change.

IISD is in the business of promoting change towards sustainable development. As a policy research institute dedicated to effective communication of our findings, we engage decision-makers in government, business, NGOs and other sectors in the development and implementation of policies that are simultaneously beneficial to the global economy, the global environment and to social well-being.

In the pursuit of sustainable development, we promote open and effective international negotiation processes. And we believe fervently in the importance of building our own institutional capacity while helping our partner organizations in the developing world to excel.
This much was picked up by David Brewer in Climate Audit comments and by Ben Pile in a guest post in Roger Pielke Jr's blog on 20 January. Pile remarked that this was another case of "laundered literature". Once again, the IPCC was relying on a document sponsored by an advocacy group, neither a scientific publication nor peer reviewed.

That, however, is as far as Pile took it, but further investigations have brought to the surface many more anomalies, which elevate the IPCC's claim from a mild curiosity to another major flaw on the scale of "Glaciergate".

Major "anomalies"

Not least of the anomalies is that the author, Ali Agoumi, is not a climate scientist, as such. Although he seems to have worked for Morocco's Ministry of Land-use Management, Water and the Environment, he currently seems to make his living from drawing up carbon credit applications under the UN's clean development mechanism. He has worked as consultant for the firm Ecosecurities, a company which specialises in carbon trading.

Next, there is no primary research in Agoumi's work – it is a review document. Furthermore, it applies only to three countries: Algeria, Morocco and Tunisia, while the IPCC report applies generically to the whole of Africa.

Agoumi's claims

As to the substance of his report, Agoumi makes a series of claims in respect of the three countries on which he reports.

Firstly (on page 4 of the 14-page document), he tells us that agriculture is primarily rain-based and is subject to climatic hazards, with "bullet points" which then assert:
  • There is strong soil erosion with extensive soil degradation. Decreasing rain-based agricultural yields with grain yields reduced by up to 50 per cent in periods of drought.

  • There is increased water needs for irrigation and decreased agricultural production due to lack of water in recent years.
  • Notably, here the statement is highly qualified. We see the reference to periods of drought and the reference to the 50 percent drop in yields applies only to grain crops. However, Agoumi expands on these assertions on page 5, telling us:
    Studies on the future of vital agriculture in the region have shown the following risks, which are linked to climate change:

  • greater erosion, leading to widespread soil degradation;

  • deficient yields from rain-based agriculture of up to 50 per cent during the 2000–2020 period;

  • reduced crop growth period;

  • risk of non-dormancy of some arboreal species;

  • reduced agricultural activity in coastal zones due to anticipated salinization of ground water;

  • reduced agricultural production linked to higher water demand in this sector, combined with an anticipated decrease.

  • These studies resulted in recommendations to adapt this sector to climatic changes, by incorporating better use of water for agriculture (new techniques), adoption of drought-resistant varieties of crops, better selection of planting dates, and supplemental irrigation in zones that lend themselves to that practice.
    Here, then, we find the unqualified reference to 50 percent deficient yields, which is presumably the passage to which the IPCC referred, in the Synthesis Report converting the time band of 2000-2020 to the single "2020". That band is interesting. Presented in a paper published in 2003 and then in the 2007 IPCC report, the claim could have been verifiable by observation.

    Confirming that the paper is a secondary source, Agoumi offers two references for the "studies" he relies upon:
    (1) Vulnerability studies on three North African countries (Algeria, Morocco and Tunisia) with respect to climatic changes, performed within the framework of the UNEP-GEF Project RAB94G31: 2000–2001.

    (2) Initial national communications by three countries (Algeria, Morocco, and Tunisia), presented at COP-7 in October 2001. These communications are available at the Web site of the United Nations Framework Convention on Climate Change (UNFCCC).
    The references

    The report of the vulnerability studies does not seem to be available online, although details of the original project can be obtained. Approved in 1994, it was intended to explore: "Building Capacity in the Maghreb to Respond to the Challenges and Opportunities Created by National Response to the Framework Convention on Climate Change."

    The purpose was to identify and strengthen "capacities of selected regional and sub-regional organizations." It "develops indigenous capacity to independently identify cost-effective greenhouse gas strategies, seeks to leverage private sector investments, and builds on national programmes to develop data of GHG sources and sinks."

    Given the focus of the project, it seems unlikely that it could offer any useful information on the effect of climate change. Furthermore, a 72-page proposal document suggests that climate studies are not part of the project.

    As to the "initial national communications by three countries" the UNFCCC site records that all three countries tabled submissions.

    The report from the Moroccan government is quite explicit, and seems to lend some support to Agoumi. It states that it expects by 2020, "a decrease in cereal yields by 50% in dry years and 10% in normal years." In other words, the first of the two statements by Agoumi is reasonably close to the mark (insofar as it applies to Morocco), where he reports: "Decreasing rain-based agricultural yields with grain yields reduced by up to 50 per cent in periods of drought."

    This, though, is just one country, and the claim seems rather tenuous - an assertion unsupported by any referenced research. The report from Algeria, however, paints a very different picture. Available in French only, on page 85 we see:
    le rapport agriculture : "a new opportunity for growth", élaboré en 1989 par la Banque Mondiale, et relatif à l'Algérie, indique que le taux de croissance annuel moyen de la production agricole doit être de 5,5 %. En admettant ce taux de croissance, on peut déduire que la production agricole devra plus que doubler à l'horizon 2020.
    The crucial sentence is the last, which states that on the basis of the assumed growth rate, "we can conclude that agricultural production will more than double by 2020." Later in the report we see that the optimal yield for the 2020 season is estimated at "nearly 30%" [more] than obtained during the agricultural year 1995-1996."

    Against that baseline, the Algerian government then ran a modelling exercise based on the temperature scenarios set out in the IPCC's third assessment report, estimating that it would see "an average reduction of grain yield of about 5.5 to 6.8% corresponding to the impacts caused mainly to climate change." Thus, in reality, its expectation is of a significant increase in production, less 5.5-6.8 percent attributable to climate change.

    Tunisia also paints a different picture. It does not offer any figures for agricultural yield losses, but notes that the country is already in "a hydrous stress situation close to a shortage". But, by modifying the evaporation and precipitation rate, it says, "the global warming will probably affect the hydrous climate balance and therefore the Tunisian water resources."

    In this way, it continues, "... the intensification of the evaporation can lead to a possible important increase of the rain falls," although it then states: "it might not be sufficient to offset the decrease of the sweet water resources." In effect, the picture is mixed but rainfall could actually increase. Rain-fed yields might similarly increase, possibly at the expense of irrigated crops.

    Therefore, Agoumi's primary references – which would have qualified as acceptable for the IPCC report – offer a mixed picture from the three countries examined. At worst, we get a 10-50 percent fall in cereal yield, the greater fall occurring only in periods of drought. Alternatively, we see a 5.5-6.8 percent trimmed from what could be a doubling of yields and then, in the third country, rainfall could actually increase – possibly (but not necessarily) improving yields of rain-fed crops.

    Returning now to Dr Pauchauri's Synthesis Report and his claim that "by 2020, in some countries, yields from rain-fed agriculture could be reduced by up to 50%," the only support he has, on the basis of the primary references, is data from one country – singular, rather than plural, unsupported by peer reviewed research. And that is set against increased production in a neighbouring country – albeit slightly trimmed - while another could actually see rain-fed yields increase overall. Then, the data apply to cereal yields only, not crops in general as is implied by the IPCC.

    Clearly though, "in one country, cereal yields from rain-fed agriculture could be reduced by up to 50% during periods of drought" would not have quite the impact of the IPCC's current claim. But using the phrasing "up to 50% reduction" is on a par with a department store reducing the price of one stock item by half and the others by considerably less, then advertising its wares as "up to 50% off". Strictly speaking, it is not untrue, but it is sharp practice. Science, it is not.

    The virus in the system

    Now the "virus" is in the system, however, it has been adopted as received wisdom, as was indeed the intention. Thus we see the BBC website tell us:

    Projected reductions in the area suitable for growing crops, and in the length of the growing season, are likely to produce an increased risk of hunger. In some countries, yields from rain-fed agriculture could be reduced by up to 50% by 2020.

    A variation is offered on the government's DFID site, without reference, gravely informing us that, in Africa, "by 2020 75-250 million people will be exposed to increased water stress due to climate change. In some countries yields from rain-fed agriculture could be reduced by up to 50%."

    The worst offender though has been Dr Pachauri himself. He has used the claim frequently in speeches and articles, for instance at the launch of the Synthesis Report on 17 November 2007 in Valencia, Spain. Then, he told the press corps that in Africa by 2020, "in some countries, yields from rain-fed agriculture would be reduced by 50%".

    At a lecture to the Royal Academy of Engineering in London on 3 October 2007, using a "powerpoint" presentation, he had a slide headed: "Expected impacts on poor regions", proclaiming:
    Malnutrition further exacerbated by reduced length of growing season in Sahelian region of Africa. In some countries, yields from rain-fed agriculture could be reduced by up to 50% by 2020.
    In another instance, he used the claim on 22 April 2008 in an interview with Reuters, embellishing the IPCC wording by warning: "Farmers who depend on rain-fed agriculture were likely to see declines of 50 percent in their yields by 2020."

    Then, on 23 December 2009, Pachauri was telling the Credit Suisse house magazine that, "... there is now growing evidence that, with climate change, the yields of several crops in different parts of the world are going down."

    Particularly vulnerable in this respect, he said, "would be rainfed agriculture, and there are parts of Africa which may see a decline of 50 per cent in agricultural yields as early as 2020."


    Such certainty had drifted a very, very long way from the science, in an area where even the experts disagree on the impacts and reject any idea of certainty. As early as 2004 – long before Pachauri had drafted his synthesis report - a group of experts had warned that climate models "have significant systematic errors in and around Africa" and that "the absence of realistic variability in the Sahel in most 20th-century simulations casts some doubt on the reliability of coupled models in this region." They concluded:
    The extent to which current regional models can successfully downscale precipitation over Africa is unclear, and limitations of empirical downscaling results for Africa are not fully understood.
    Even during the review process of the Working Group II report, scientists were posting warnings. Richard Anyah, from Rutgers University, told the authors, "It should be noted that the uncertainty in precipitation projections over Africa still makes it difficult to assess future vulnerability to food insecurity given the over-dependence on rain-fed agriculture."

    Furthermore, concern about the uncertainties was widespread. The Met Office Hadley Centre, experimenting with new climate models, reported in 2004 that while summer rainfall in Africa "is most likely to decrease" the wide spread of the result "shows the high degree of uncertainty in this prediction."

    Possibly reflecting that uncertainty, the Stern Review in May 2006 offered the view that "small rises in temperature and reductions in rainfall could 'tip the balance' and lead to severe water shortages and reductions in crop yields". These could fall, the report stated, not by the IPCC's 50 percent by 2020, but a more conservative amount. It predicted falls "by as much as 30 per cent by the 2050s."

    While Pachauri sticks to his flawed mantra, expert opinion diverges even further. In October 2009, the Grantham Institute for Climate Change published a paper by Professor Gordon Conway. It re-emphasised that "there is much that we do not know" about the effects of climate change in Africa.

    "Will the overall fall in agricultural production be very large or relatively small?" Conway asked, noting that: "Part of our ignorance comes from a growing but, nevertheless, poor understanding of the drivers of the African climate and their complex interactions."

    "Part is due to a severe lack of local weather data, particularly for central Africa," he continued. "This lack of knowledge makes it difficult to validate climate models and hence predict with any degree of accuracy what will happen as a result of climate change at a country, or even sub-regional level in Africa."

    What applies to climate models also applies to crop yield estimates. In the same month as Professor Conway's report, the prestigious International Food Policy Research Institute updated one of its reports, entitled "Climate Change: Impact on Agriculture and Costs of Adaptation".

    Applying a series of sophisticated computer models to estimate crop yields in Africa, it took the worst-case scenario in Sub-Saharan Africa and, even there, reported "mixed results". It forecast "small declines or increases in maize yields and large negative effects on rainfed wheat."

    Actual yields would be affected not only by climate directly but by "autonomous adaptation as farmers respond to changing prices with changes in crop mix and input use." With those issues factored in, the institute estimated that, by 2050, the negative effects of climate change on crop production in Sub-Saharan Africa, where they will be "especially pronounced", amounted to a 15 percent decline in rice production, 34 percent in wheat and 10 percent in maize.

    This picture, set 30 years later than the 2020 timeframe used by the IPCC, offers a more realistic insight into the complexities of agriculture and the response of farmers to changing circumstances. Nothing of that is conveyed by the apocryphal vision of the IPCC.

    The end game?

    Only last week, the government's chief scientist, Professor John Beddington, warned that scientists must be more "honest and open" about the uncertainties of global warming.

    As we write, Professor Chris Field, currently lead author of WGII - speaking from snow-bound Washington - admits that he is "not aware" of the support for the statement in the Synthesis Report, defending the Chapter 9 statement only insofar as he suggests that it refers to climate change "exacerbating current risks". He makes no attempt to defend the 50 percent figure or the time scale offered by the IPCC.

    Even the mildest critics of the IPCC and Dr Pachauri might now be moved to observe that they have eschewed uncertainty, to project the most pessimistic scenario imaginable – with no scientific support and a great deal of embellishment. After "Climategate", "Glaciergate", "Amazongate" and now "Africagate" can either survive?