The quality and value of an area depend primarily on what is found within its designated boundaries. To determine the value of such areas in terms of boundaries, we have developed a methodology for integrating multi-attribute utility functions with spatial analysis so that desirable, or appropriate, boundaries can be determined on the basis of the goals and objectives for the park. The key to successful evaluation is the development of measures for what decision makers value, rather than merely using what is easily measured. In a detailed example for planning of national parks, the fundamental objectives are conservation and societal use of the park resources. These are divided into sub-objectives, and then functional objectives, which are measured with a family of spatial analysis functions. Modeling is iterative, so that available spatial-analysis techniques may be used to produce acceptable measures from available information. The resulting value model is a key component in an integrated decisionsupport system that allows the decision maker to evaluate alternative plans before selecting one of them. Mathematical and other definitions used in structuring this problem are generalizable to other geographic decision support applications.