Improving the effectiveness of environmental management decisions with optimization and uncertainty analysis techniques

The stochastic nature of environmental processes and forcing functions diminishes the credibility of management decisions that are based on models that employ ''average'' conditions. Our thesis is that the environmental decision making process must quantitatively account for environmental uncertainties. In doing so, management strategies can be formulated that are based on acceptable levels of risk to the environment. We have extended this uncertainty analysis approach to include a linear programming optimization algorithm in order to demonstrate that acceptable risk scenarios can be devised which are cost effective as well. We demonstrate this point for the Great Lakes by showing that present and proposed phosphorus management plans have neither the probability of success nor the cost effectiveness of a management plan that is based on a combined uncertainty-optimization analysis. Present and proposed phosphorus management plans are predicted to have at most a 55% overall probability of meeting Great Lakes phosphorus goals. The combined uncertainty-optimization approach points to management options whose overall probability of success in meeting Great Lakes phosphorus goals varies between 55% and 99%. The costs associated with these options also vary, but savings as great as 400 million dollars per year could be realized with no diminishment in goalmore » achievement when compared to present and proposed management plans.« less