• Ãlvarez, Milagros
  The Graduate School, University of Maine
  

This research proposes the use of Euclidean distances as a decision support tool in forest ecosystem management in a framework of analysis that integrates linear programming, growth and yield simulation software and Geographic Information Systems. The developed methodology, which integrates economic, ecological and ecological values, helps decision makers better understand the implications of their decisions and the trade offs that occur when forest values compete. The study also tests the hypothesis that forest management directions that favor the greatest variety of conditions lead to a greater aggregate value than those directions that favor narrower goals.

The study area is composed of more than 36,000 acres of State-owned land in 'Western Maine. The dissertation is organized in six parts. The first two parts reviewed definitions of forest values and existing quantifiable methodologies that estimate these values. To provide management guidance for recreational opportunities, the third part analyzed recreational supply and demand at the local and state level. This analysis led to the conclusion that the area should retain its remote and undeveloped character while providing primitive and semi-primitive recreational opportunities. Parts four and five created a modeling environment that allowed the simulation of M management scenarios varying from "high intensity management" to "no management," and integrated a variety of computer applications including ARC/INFO@, Forest Vegetation Simulator, and Spectrum. An evaluation of the capabilities and limits of the software used revealed that their integration represents a powerful tool in forest management. The last part presented a new methodology of analysis. The researcher created a nine-dimension space where each axis represents the percent decrease of each analyzed outcome relative to the maximum capacity of the forest to produce a benefit in the absence of any other competitive uses. The Euclidean distance in the defined nine-dimension space quantified how far each simulated scenario was from the theoretical optimum. This distance represented a comparative measurement across scenarios and was compared to the variety of benefits provided by each scenario in order to test the original hypothesis. The researcher concluded that Euclidean distance represents a simple, flexible and accurate quantitative indicator of economic, social and ecological values of any management plan given any number of feasible, desired goals.