Adaptive-decision strategy offsets uncertainties in climate sensitivity

by University of Illinois Urbana-Champaign

he uncertainty of climate change be cause of global warming is much greater than previously thought. As a result, policy-makers should adopt a robust, adaptive-decision strategy to cope with potential consequences, researchers at the University of Illinois say.

    As reported in the Oct. 16 issue of the Journal of Geophysical Research, UI atmospheric scientists Natalia Andronova and Michael Schlesinger found there is a 54 percent chance that climate sensitivity lies outside the 1.5 to 4.5 degree Centigrade range announced by the Intergovernmental Panel on Climate Change.

    “This is definitely not good news,” Schlesinger said. “If the climate sensitivity is greater than the IPCC's upper bound, climate change could be one of humanity's most severe problems of the 21st century. If, however, it is less than the lower bound, then climate change may not be a serious problem for humanity.”

    In a study supported by the National Science Foundation and the US Department of Energy, the researchers used a simple climate/ocean model and the observed near-surface temperature record to estimate the probability distribution for climate sensitivity. For each of 16 radiative-forcing models, the changes in global-mean near-surface temperature were calculated for the years 1765 through 1997. The radiative-forcing models included greenhouse gases, anthropogenic sulfate aerosol, volcanoes and solar irradiance. The researchers found that, as a result of natural variability and uncertainty in the radiative forcing, it is 90 percent likely that climate sensitivity lies between 1 and 10 degrees Centigrade.

    While some scientists have argued that the IPCC should assign subjective probabilities to its various scenarios of future climate change, Schlesinger and colleague Robert Lempert (of the Rand Corp.) disagree. Optimization plans should not be based upon subjective probabilities because “decision-makers must form a climate policy acceptable to groups with many different, yet plausible, estimates of the likelihood of alternative futures,” the two researchers wrote in the July 26 issue of the journal Nature.

    The large uncertainties associated with the climate-change problem can make conventional policy prescriptions unreliable, Schlesinger said. “It could take a fair fraction of a century to acquire enough observations to significantly reduce the level of uncertainty, and by then it may be too late to do anything about it. By using an adaptive-decision strategy, however, we can observe the damages due to climate change, and the rate of change of the cost differential between fossil fuels and non-fossil fuels. Depending upon what we see, we can alter what we do.”

    By objectively estimating the likelihood of the climate sensitivity having any particular value - that is, by its probability distribution - the crafting of robust, adaptive climate-change policy could be greatly facilitated, Schlesinger said. “Such a strategy could also aid in the negotiation process, because decision-makers could make more reasonable and defensible choices about climate-change policy without requiring highly accurate or widely accepted predictions of the future.”

    Contact: University of Illinois at Urbana-Champaign James E. Kloeppel, Physical Sciences Editor (217) 244-1073;