Issues affecting so-called 'global warming' - Position paper for U.S. Climate Change Science Program Workshop (Dec 3-5, 2002)
The primary purpose of this Position Paper is to define the crucial climate research needed to resolve major scientific disparities. In the process, it will become clear that one can draw conclusions from the body of climate data assembled within the IPCC Science report (Third Assessment, 2001)  that are substantially different from the conclusions exhibited in their current Summary for Policymakers (SPM).
We also try to suggest priorities for the research to be sponsored by the USGCRP by pointing to broad areas that bear on public policy, where there are starkly different interpretations:
Areas for Research
1. The SPM states that the climate is currently warming as a result of human activities. But the totality of available data can also be interpreted to show that the climate is not warming perceptibly, and that the human impact is currently too small to be discerned. In any case, even if there were an observed warming, one needs to look for characteristic "fingerprints" in the detailed record to assign such a warming to human causes. [As explained in the IPCC report, this would require comparing observed trends of the geographic and altitude distributions of temperatures with climate model results.]
2. The SPM claims to show that the observed temperature history of the 20th century can be well explained as the result of natural and anthropogenic climate forcings. But one can argue that this claim is really nothing more than a curve-fitting exercise with the use of adjustable parameters. [This bears on the question whether current climate models have been validated and whether they can be relied on to predict future changes in climate.]
3. The SPM claims that after a steady decline of (Northern hemisphere) mean temperatures since 1000 AD the 20th century was the warmest in the past 1000 years --- based on a single analysis of proxy data. But other analyses of the same or similar data show very different results. [This includes also analyses published in earlier IPCC reports. In any case, there is nothing special about "1000 years;" it is generally accepted that temperatures were much higher in earlier periods of the Holocene.]
4. The SPM predicts significant deleterious impacts from a future warmer climate. In particular, it links the observed rise in global sea level during the 20th century to the warming trend during the same century. But the logic of this argument is not compelling, and one can adduce data from the IPCC report that contradict this facile conclusion.
5. Our understanding of the natural and anthropogenic forcing of the climate is still evolving. This becomes quite evident when one compares the three IPCC Assessment reports of 1990, 1995, and 2001. In particular, we are learning about the strong influence of variability in solar activity on decadal-timescale fluctuations of climate. New results are also forthcoming on both natural and anthropogenic aerosols, their temporal and geographic variability, and their effects on climate. Our understanding of the past and present trends in the atmospheric concentration of greenhouse gases is still poorly understood. In general, therefore, prediction of future forcing is pretty much a matter of speculation, especially when one factors in volcanic eruptions and El Nino events. [This scientific uncertainty is distinct from the uncertainty in predicting GH gas emission scenarios based on demographic-economic projections.]
6. The key issue is validation of the climate models (GCMs), without which, one cannot make reliable forecasts of future climate - even if one knew the future values of all radiative-forcing parameters. Validation means demonstrating whether current climate models are capable of accounting for observed temperature patterns; one needs to compare the "fingerprints" in temperature and other climate parameters with observations. There are still major uncertainties about the physics going into the models, particularly about internal feedbacks from clouds and distribution of water vapor.
Because of the inter-connectedness of all changes in the condition of the atmosphere and oceans, and even of the cryosphere, lithosphere, and biosphere, one can argue that all studies are of potential importance. Nevertheless, if priorities were to be assigned, we would choose the following:
1. The single most important research issue revolves around the question of whether the global mean climate is currently warming [since 1979]. As confirmed in the report by the National Academy of Sciences /National Research Council , surface data since 1979 show a warming trend but atmospheric temperatures measured by satellites and (independently) by weather-balloon radiosondes show no such warming,
2. Next in importance might be studies that compare "fingerprints" derived from climate models with recent observations of the geographic and altitude distribution of temperature trends. [This bears on the question whether current climate models have been validated and whether they can be relied on to predict future changes in climate.]
3. Improvements in the physics of climate models are needed, especially. the uncertainties about the magnitude and sign of feedbacks from clouds and water vapor.
4. Of the various impacts of a postulated global warming, sea level rise may be considered of special importance and deserves a greater priority than other impacts
5. The other studies listed above, while of great scientific interest, probably carry a lower priority, and have less policy significance. In this category also fall studies of natural climate forcing to account for pre-historic terrestrial climate changes, including ice ages, and climate changes on other planets.
1. UN-IPCC. Climate Change 2001: The Scientific Basis. Cambridge University Press, Cambridge, 2001. Differences between the IPCC report and the SPM are exhibited in a separate SEPP study.
2. National Research Council. Reconciling Observations of Global
Temperature Change. National Academy Press. Washington, DC. Jan. 2000.
Furthermore, there are significant differences between different analyses
of the surface data, as well as other problems with the corrections
applied to these data. These are exhibited in a separate SEPP study.