In order to estimate the impacts of a future climate change it is of course necessary to know many details of that change. All current methods for supplying those details fall short of what is needed, and this deficiency presents a large challenge to the climate science community.
Future climates depend on processes within and interactions among three distinguishable but coupled systems--human society, the physical climate system, and the biosphere external to human society. Important interactions take many forms. The rate at which society emits heat-trapping gases, for example, depends on policy actions at the national and international level, actions likely to be shaped in part by knowledge of possible rates of climate change and the impacts that will follow. The climate and the biosphere are coupled through changes in atmospheric composition, surface color and roughness, evapotranspiration, and other factors. Although most studies focus on one of these systems, an increasing number of investigators now consider one or more interactions among the systems. See also Dickinson, Land Atmosphere Interaction, in this issue.
The favored approach today for estimating future climate change in the face of the uncertainties consist of 1) accepting a wide range of Infrared-trapping gas emission scenarios; 2) with the help of ocean and ecosystem models, converting each emission scenarios into a scenario of atmospheric concentration of the relevant gases; and 3) using as complete a numerical model of the climate as available to convert these concentration scenarios into patterns of future climate change. Each step will be reviewed in turn.