Resource Management, Decision Models in

There are a wide variety of natural resource management issues that are addressed using a suite of different management tools. Primary among these are decision models. Natural resource management problems include developing structures and procedures to provide reliable water supply for domestic and industrial use, irrigation, recreation, and flood control; ensuring that productive wetlands are preserved and maintained; providing conservation lands to ensure maintenance of biodiversity; managing air and water quality; and fisheries management, among many others. A wide variety of decision models are also available to address these issues, ranging from simple sequential pairwise comparisons of options to more sophisticated models requiring vast amounts of data and computing power. Decision models help generate plausible alternatives for action and help choose among these alternatives. There are also many ways to categorize these decision models; Figure 1 provides a brief typology.

A decision problem for natural resource management can consist of the overall goals (objectives) for wise management—for example, providing a “safe yield” of a renewable resource to ensure its sustainability or efficiently using a limited natural resource such as fossil energy. The discrete choices to be made are the alternatives (e.g., which forest stands to harvest, choosing among multiple reservoir projects, choice of fuels for generating electricity). For each of the alternatives, several attributes may be important for the decision (e.g., in the case of conservation lands, this may be their cost, location, and habitat value for different species). The discrete choice models in Figure 1 differ as to whether a single criterion or multiple criteria are used and if time sequencing and uncertainty are major factors.

Cost-benefit analysis has been long and widely used, particularly for large-scale, publicly funded projects in water resource management and in environmental regulatory decisions. When choosing among alternative projects, procedures, or standards, the monetary costs and benefits for all attributes of each alternative are established. The choice is based on the maximization of a single objective: economic efficiency. This requires choosing the alternative whose net benefits (total benefits–total costs) are the largest. To account for a time sequence of costs and benefits, discounting (valuing all benefits and costs in current dollars) is used. A difficulty in the use of this technique involves the valuation of all attributes for all alternatives, especially benefits, particularly when they are not traded in the markets.

Multicriteria evaluation (MCE) decision models have been used when more than one criterion (objective) is employed and/or when the attributes for each alternative can be measured in their natural units, obviating the need for strict valuation. often, the measures of the attributes for each of the alternatives are standardized. This is usually linearly scaled from 0 to 1 or 1 to 100, or through the use of value functions or fuzzy set functions, with the purpose of allowing fair comparison among the measures. The criteria (or objectives) are then weighted using the decision makers’ preferences. A weighted linear combination is obtained for each alternative by summing the product of the criteria importance weight and the alternatives value on each of the attributes; the decision rule then is to choose the alternative having the largest value. These weights can be elicited directly by simply assigning weights, [Page 2452][Page 2453]through structured transitive comparisons (multi-attribute utility theory, or MAUT), or by making a series of pairwise comparisons of the alternatives (the analytic hierarchy process, or AHP).

...