Decision support system for sustainable groundwater management

The major goal of this research is to develop a Decision Support System (DSS) to aid decision-makers in evaluating groundwater pumping and recharge policies in semi- and areas, with particular emphasis on environmental objectives. For decades, groundwater simulation models have been linked to optimization in order to build management tools. The general assumption is that there exists a single, centralized decision-making authority that defines the management objectives. However, when a single decision-maker is absent and instead replaced by multiple stakeholders, definition of management objectives is often subject to imprecision. Such imprecision is not due to randomness but rather because of factors such as ambiguity, generality or vagueness, stemming from the use of linguistic variables to describe the states of the system. Thus one of the goals of this research is to explore the relationship between objective imprecision and the required reliability of a simulation model. The proposed methodology combines multiobjective optimization and multicriteria programming to generate a decision-making framework for groundwater management. The DSS consists of a multiobjective optimization model that considers three objectives: minimizing costs, maximizing aquifer yield, and minimizing drawdown at selected locations. A fuzzy decision-making model searches for the non-dominated set of solutions that are acceptable by a group of stakeholders. Thus the methodology combines three non-fuzzy objectives and one fuzzy objective to respectively define the search space and to evaluate the compliance of the solutions with additional environmental criteria that may be subject to imprecision because of the multidisciplinary nature of environmental issues. Once the decision-making framework is established, the next step is to evaluate the reliability of the groundwater simulation model with respect to management applications. A first-order analysis translates model parameter uncertainty to the decision and solution space so that acceptable ranges of parameter uncertainty can be obtained by solving a nonlinear, nonconvex optimization problem, given acceptable uncertainty in the decision space. The methodology is being applied to a real case, the San Pedro River Basin. The Basin is located in southeastern Arizona and represents an ideal case study for sustainable water resources management studies because of its size, its ecological relevance, and the level of stakeholder organization that has been reached in the Basin.