Identification of Tolerable Risk Levels for the Operation of a Multi-Purpose Reservoir in Northern California

Abstract:

Ensemble Forecast Operations (EFO) is a risk-based approach to reservoir flood control operations that incorporates ensemble streamflow predictions (ESPs) generated by the California Nevada River Forecast Center (CNRFC) using the National Weather Service Hydrologic Ensemble Forecast Service (HEFS). Reservoir operations for each member of an ESP are individually modeled to forecast system conditions and calculate risk of reaching critical operational thresholds. EFO was developed for Lake Mendocino, a 111,000 acre-foot reservoir near Ukiah, California, to evaluate its viability to improve reservoir management, and provide decision support to reservoir operators in making releases ahead of major runoff events. Tolerable risk levels were identified for proof-of-concept testing and evaluation of EFO for Lake Mendocino, which involved simulating operations under numerous risk curves and evaluating the performance with each curve as measured by an objective function. The objective function was developed through stakeholder engagement and based on the project objectives to maximize storage reliability, minimize occurrences of downstream flooding, and avoid emergency spillway releases. EFO was simulated using a 26-year (1985-2010) ESP hindcast of the HEFS generated by the CNRFC, which provides 61-member daily ensembles of 15-day flow forecasts. Utilizing an optimized risk tolerance curve, EFO yields a 33% increase in median storage at the end of the flood management season (May 10) over existing operations without marked changes in flood frequency for locations downstream from Lake Mendocino. These results warrant further research in EFO for Lake Mendocino and further refinement of risk tolerance levels for improvement of reservoir operations to meet management objectives. EFO framework could also be applied to other multi-purpose reservoir systems using similar methods to identify tolerable risk levels.