Use of Probability-Based Streamflow-Depletion Criteria to Constrain Optimal Ground-Water Withdrawals

Increased demands for water to meet human and ecosystem requirements are creating a need for improved methods for planning reliable water supplies while meeting environmental-protection goals. Streamflow depletion caused by ground-water withdrawals can be an environmental problem when such depletions reduce streamflow below minimum levels required to sustain healthy ecosystems. The approach to water-supply management taken in this work is to use probability based streamflow-depletion criteria that account for natural variability in streamflow, as quantified by monthly streamflow-duration curves. In this approach, ground-water withdrawal schedules are determined using simulation-optimization models that maximize ground-water withdrawals while simultaneously meeting maximum allowable streamflow-depletion rates that are based on specified monthly low-flow probabilities. The resulting optimal withdrawal schedules provide water suppliers with an assured maximum supply of water each month for a given level of maximum streamflow-depletion risk. For example, the water supplier and environmental regulators may decide to allow streamflow to be reduced to some percentage of the natural period-of-record minimum monthly one-day streamflow for each month of the year. These minimum monthly one-day streamflows correspond approximately to a one-percent probability that the minimum monthly streamflow criteria will actually be realized over a long period of withdrawal. The optimal withdrawal schedules determined for a given constraint set will result in new monthly streamflow-duration curves that are shifted downward relative to the natural streamflow-duration curves but are constrained to lie above each of the minimum monthly required flows.