This study investigates the operational decision process for Okanagan Lake, a natural lake regulated by a dam at the outlet for flood control, irrigation and water supply purposes. In addition, the Lake supports a substantial tourist industry. The Lake is principally supplied by snowmelt and a forecast of total inflow volume during the critical runoff season is available to assist the operator. The operational decision process was found to differ from the sequential decision basis of many Operations Research techniques and the absence of information on costs and benefits precluded the use of conventional optimization procedures. The importance of making the best use of the inflow forecast to achieve the operational goals was recognized and was used as the basis of the decision analysis developed. The method developed assesses possible immediate operational decisions by evaluating the effectiveness of future discharges to correct for past decision errors. The evaluation is made in terms of the probabilities of exceeding Operational constraints and of achieving operational goals. The method involves simulation of sets of monthly inflows for the remainder of the runoff season given an inflow volume forecast and knowledge of the probable accuracy of the forecast; computation of water levels which would occur with various operating procedures; frequency analysis of the resulting levels; interpretation of the frequencies as probabilities; and presentation of the resulting information describing the operational situation in readily assimilable form. / Applied Science, Faculty of / Civil Engineering, Department of / Graduate