IPCC Global Warming Report Fails to Deliver Sound Policy

Climate change certainty misrepresented

Climate is in the news again, as the United Nations Intergovernmental Panel on Climate Change (IPCC) has just released the Summary to Volume 2 of its Third Assessment Report (TAR) on climate change.

The IPCC released Volume 1 of the TAR in December of 2000, summarizing what is known about the science of climate change and speculating on the broad elements of how the climate is expected to change in the future. Volume 2 picks up where Volume 1 left off, translating estimated future temperatures into estimated future weather conditions, out to the year 2100.

The news from Volume 2 is gloomy. The Volume 2 “Summary for Policymakers” raises the specter of coastal inundation, increasingly violent weather, more droughts, increased spread of mosquito-borne illnesses, crop failures, and more. Even more depressing than the predictions, and the uncritical way in which the media are accepting them, is the way in which the summary puts forward a distorted image of the actual state of knowledge regarding future climate change impacts.

Global warming, and the climate changes that may result from it, is a source of concern to many people, and rightfully so. Radical changes in climate could cause severe disruption to ecosystems and human societies, regardless of the underlying causes of that change.

But rational policymaking requires an understanding of climate issues that reflects the true state of knowledge, not biased representations used to justify pre-selected policy objectives. The IPCC report process is a tremendous effort involving many hundreds of well-intentioned scientists. The three TAR volumes, which run to 1,000 pages each, are rich in detail and are careful to include numerous caveats that accurately portray the large uncertainties in predictions of future climate changes and impacts. However, the “Summary for Policymakers” in the new Volume 2, like the summary from the Volume 1, fails to carry that careful portrayal forward. Instead, the summary misleads more than it informs, passing off speculation as fact, and blurring key distinctions needed for sound policymaking.

False Certainty

Press reports have trumpeted the higher “degree of certainty” that this new TAR volume represents. One Associated Press article from February 18, 2001 characterized Africa’s future as follows:

Grain yields are expected to decrease, and there will be less water available. Desertification will be worsened by reductions in average annual rainfall, especially in southern, North and West Africa. Coastal settlements in Nigeria, Senegal, Gambia, Egypt, and along the East-Southern African coast will be hit by rising sea levels and coastal erosion. [emphasis added]

The Volume 2 summary itself seems to lose track of the difference between predictions, and certainties. As the summary says regarding coastal zones and marine ecosystems,

Many coastal areas will experience increased levels of flooding, accelerated erosion, loss of wetlands and mangroves, and seawater intrusion into fresh water sources as a result of climate change. The extent and severity of storm impacts, including storm-surge floods and shore erosion will increase as a result of climate change including sea-level rise. High latitude coasts will experience added impacts related to higher wave energy and permafrost degradation. Changes in relative sea level will vary locally due to uplift and subsidence caused by other factors.

But footnote number six in the summary reveals the subjective nature of the report’s predictions. Though the degree of certainty is reported through numerical ranges, the actual determination of certainty is a subjective exercise in which the authors of the report simply state their opinion and assign a number to it. The analysis does not include any mathematical tests of certainty, nor does it attempt to derive statistical confidence intervals. Instead, the degree of certainty of the report’s predictions is based merely on a self-polling of the authors’ professional judgment. As the footnote explains:

In this Summary for Policymakers, the following words have been used where appropriate to indicate judgmental estimates of confidence (based upon the collective judgment of the authors using the observational evidence, modeling results, and theory that they have examined): very high (95% or greater), high (67-95%), medium (33-67 %), low (5-33 %), and very low (5% or less). In other instances, a qualitative scale to gauge the level of scientific understanding is used: well established, established-but-incomplete, competing explanations, and speculative.

One has to applaud this attempt to quantify confidence levels, but one also has to ask whether these “confidence” measures are inherently misleading. Some expert comments about the models used to predict future regional climate changes suggest so. According to a Science news brief from June 2000, Jerry Mahlman, Director of NOAA’s Geophysical Fluid Dynamics Laboratory observes that when regional climate models try to incorporate external factors such as population and economic growth rates, “the details of future climate recede toward unintelligibility.” Climate modeler Filippo Giorgi, of the Abdus Salam International Center for Theoretical Physics, explains, regarding regional climate models of the sort used by the TAR authors, that “For the most part, these sorts of models give a warning, but they tend to give very different predictions, especially at the regional level, and there’s no way to say one should be believed over another.”

Further, as climate modelers from four separate climate modeling centers observe in an article from Nature, “Forecasts of climate change are inevitably uncertain.” They go on to explain that “A basic problem with all such predictions to date has been the difficulty of providing any systematic estimate of uncertainty,” a problem that stems from the fact that “these models do not necessarily span the full range of known climate system behaviour” (Allen, et al., October 2000).

What About the Human Factor?

The Volume 2 summary discusses the many things that can happen as a result of climate change, but a key distinction is buried in the very first footnote, which explains that the climate change talked about in the report includes all forms of climate change, whether it is warming or cooling, human-caused, or non-human-caused. The summary does not identify the relative contribution of human activity to overall global warming.

However, it is precisely this information that is necessary for making determinations of appropriate action – whether to try and reduce a particular human activity in order to forestall global warming, or whether to take actions to ensure that ecosystems and human systems are prepared to adapt to climate change, whatever the cause.

It isn’t surprising that the media have not picked up on this, since the Volume 1 summary also glossed over the subject of climate change caused by human activity. Climate impacts are calculated based on predicted increases in global average temperature, using models that break out regional impacts of such changes. But predicted increases in temperature taken from the first summary were fed into “impact models” without observing that the majority of physically observed warming since 1860 was of non-human origin, and would have happened regardless of human action. The Volume 2 Summary similarly fails to point out what portion of its predicted future impacts flow from such non-human climate influences. As a result, the blurring of the first report carries over to the second, creating the incorrect impression that both climate change and its future impacts are solely the result of human activity.

This muddling of the underlying causes of climate change makes it difficult for policymakers – the very people intended to receive the report – to make intelligent decisions about whether to take actions to reduce the human contribution to climate change, or to prepare to adapt to the climate changes that are happening from forces outside of human control, such as changes in solar output.

Pessimistic and Unsubstantiated Assumptions

Another problem with the current report data is that all of its predictions are based on estimates of global warming and sea-level rise from Volume 1 – and some of those predictions were of questionable validity.

Working from more extreme “worst-case” estimates, the Volume 1 summary suggested a higher range of potential warming and sea-level rise by 2100. The average global temperature in the new report is modeled to increase from 1.5 to 5.8 degrees Centigrade (2.7 to 10.4 degrees Fahrenheit) by 2100. Predicted sea-level increases under the new scenarios range from 14 to 80 cm. (5.5 to 31.2 inches) by 2100.

However, the output of these models is only as good as what was put in. While Volume 1 competently describes what’s currently known in climate change science, it does depend on more than a few questionable assumptions. These assumptions underlie the worst-case scenarios predicted for temperature and sea-level rise, including:

  • No mid-course abatement or education programs will be implemented between now and 2100, even in the face of what are predicted to be extreme climate changes;
  • Global deforestation will not decrease;
  • World gross domestic product (GDP) will increase by a factor of 10 by 2100;
  • Most energy production will be from carbon-based fuels, with limited technology growth;
  • Carbon dioxide emissions will nearly quadruple by 2100;
  • Methane emissions will more than double by 2100;
  • Carbon monoxide emissions will nearly triple by 2100;
  • Volatile organic carbon emissions will nearly triple by 2100; and
  • Fluorocarbon levels will rise dramatically by 2100, in some cases by a factor of 100.

Volume 1 provided little justification for these assumptions, which were generated by a special working group of the IPCC and which were reviewed in an alternate, informal review process far less rigorous than the scientific review process used in either the Volume 1 or Volume 2 review processes.


While everyone is rightly concerned about prospective changes in the global climate, it’s questionable whether the TAR reports contribute to the process of developing effective climate policies. Though the underlying reports carefully hedge with statements of uncertainty, the summaries of the reports fail to accurately reflect the uncertainties that are a vital part of crafting rational policy responses.

The newly released Volume 2 “Summary for Policymakers” is no exception. It makes predictions based on simple models that: do not take into account current or historic climate phenomenon, are not calibrated to observed climate phenomenona, fail to emulate fundamental climate processes, and project a veneer of certainty that is not supportable from underlying technical reports or statements regarding similar exercises made in mainstream science journals.

Even accounting for the limitations of the models used in projecting 100-year weather patterns, the predictions of the report reflect garbage-in-garbage-out modeling. They are based on possible future scenarios that include speculative and often unreasonably pessimistic predictions about population, fuel use, technology development, international trade, development rates, and dozens of other factors.

By lumping together predictions of warming that have some human, and some non-human components, the report also fails to provide the kind of information that policymakers need in order to choose between policies that aim to slow global warming – by reducing greenhouse gas emissions – and policies that aim to adapt to global warming – by investing in technologies to offset specific climate impacts of concern, such as increased floods, or growth in insect-borne diseases.

The ramifications of climate change policy are too far-reaching to be based on distorted representations of the current state of knowledge in either climate science, or climate predictive ability. The newly released summary of IPCC’s Third Assessment Report on Impacts, Adaptation, and Vulnerability follows a pattern of policymaker summaries that weaken the link between climate policy and reality, and subvert the goal of ranking environmental policy interventions based on rigorous use of scientific principles.

Dr. Kenneth Green is senior fellow at Reason Foundation and Chief Scientist at Frasier Institute.