The Role of Models in Prediction for Decision

The processes of science and decision making share an important characteristic: success in each depends upon researchers or decision makers having some ability to anticipate the consequences of their actions. The predictive capacity of science holds great appeal for decision makers who are grappling with complex and controversial environmental issues by promising to enhance their ability to determine a need for and outcomes of alternative decisions. As a result, the very process of science can be portrayed as a positive step toward solving a policy problem. The convergence—and perhaps con-fusion—of prediction in science and prediction for policy presents a suite of hidden dangers for the conduct of science and the challenge of effective decision making. This chapter, organized as a set of inter related analytical vignettes, seeks to expose some of these hidden dangers and to recommend strategies to overcome them in the process of environmental decision making. In particular, this chapter will try to distill some of the lessons gleaned from research on modeling, prediction, and decision making in the earth and atmospheric sciences for quantitative modeling of ecosystems. One clear implication is that conventional approaches to modeling and prediction cannot simultaneously meet the needs of both science and decision making. For ecosystem science, there fortunately exists a body of experience in understanding, using, and producing predictions across the sciences on which to develop new understandings of the relationship of science and decision making to the potential benefit of both research and policy.