On an improved sub-regional water resources management representation for integration into earth system models

Human influence on the hydrologic cycle includes regulation and storage, consumptive use and overall redis- tribution of water resources in space and time. Representing these processes is essential for applications of earth system models in hydrologic and climate predictions, as well as im- pact studies at regional to global scales. Emerging large-scale research reservoir models use generic operating rules that are flexible for coupling with earth system models. Those generic operating rules have been successful in reproducing the overall regulated flow at large basin scales. This study investigates the uncertainties of the reservoir models from different implementations of the generic operating rules us- ing the complex multi-objective Columbia River Regulation System in northwestern United States as an example to un- derstand their effects on not only regulated flow but also reservoir storage and fraction of the demand that is met. Nu- merical experiments are designed to test new generic operat- ing rules that combine storage and releases targets for multi- purpose reservoirs and to compare the use of reservoir usage priorities and predictors (withdrawals vs. consumptive de- mands, as well as natural vs. regulated mean flow) for config- uring operating rules. Overall the best performing implemen- tation is with combined priorities rules (flood control storage targets and irrigation release targets) set up with mean annual natural flow and mean monthly withdrawals. The options of not accounting for groundwater withdrawals, or on the con- trary, of assuming that all remaining demand is met through groundwater extractions, are discussed.