Is Rational Decision Making Appropriate for Management of Irrigation Reservoirs?

Existing rational metrics of performance used to identify optimal reservoir management policies are examined to determine if they adequately measure the desirability of a water release strategy. It is shown that such performance metrics, while regarded as providing a rational framework for decision making, may in fact only be nominally rational and in fact at odds with the true attitudes and perceptions of the decision makers and users affected by the water releases. This situation is particularly relevant for reservoirs designed and operated to supply water to irrigation projects. A means for a more complete representation and evaluation of the possible consequences associated with a release decision is therefore necessary. A model capable of replicating the manner in which risks associated with release decisions are perceived, interpreted, and compared by a decision maker is proposed. The model is based upon neural network theory and enables the more complete representation of the risk of a particular decision to be considered in making decisions on reservoir releases.