Analysis of reservoir system reliability constrained by natural salt pollution.

摘要

A simulation computer program, RESSALT, was developed to study natural salt pollution in a river basin. RESSALT is a generalized water quantity and quality model for conservation purposes. The model was used to study the effects of natural salt pollution on the reliabilities of reservoir system operation in the Brazos River Basin.; The Brazos River Basin case study considered a system of 17 control points consisting of 13 major reservoirs and four non-reservoir control points. The system was simulated for conservation purposes. Two case studies consisting of specified water use scenarios study and yield-reliability study were considered. In each study, the system simulations were performed using 1984 and 2010 sediment conditions. For each sediment condition, alternative reservoir operating plans were simulated. The simulations were based on an 85-year historic record of monthly unregulated streamflows and salt loads. The salt constituents considered were total dissolved solids (TDS), chlorides (Cl), and sulfates ({dollar}SOsb4{dollar}).; The specified water use scenarios study involved the analyses of 18 alternative simulation runs based on water use and reservoir sedimentation condition, simulation periods, salt concentration constraints, salt load input data sets, proposed salt control dams, and model options for handling negative incrementals.; The system performance is based on volume and period reliabilities. With no specified maximum allowable salt concentrations, the 1984 water use diversions can be met with only minimum shortages while the higher demands of the 2010 water uses experienced widespread water supply shortages. Specifying allowable salt concentrations significantly reduces the diversion reliabilities of both the 1984 and 2010 water uses. The proposed salt control dams; however, improve the volume and period reliabilities of the system for the 1984 and 2010 water use scenarios.; Different alternative salt load data sets were used. The alternative salt load set that incorporates the random components to the expected salt load values reflects very well the natural variations of salinity in the river. However, the different data sets yielded comparable results.; The simulation analyses of the yield-reliability study demonstrate the sensitivity of the yield-reliability relationships to specified allowable salt concentrations, reservoir system operating strategies, and salt control dams. The result indicated that reliabilities are very sensitive to the maximum allowable salt concentrations for a specified yield level. The sensitivity, however, decreases with increasing yield levels.; Operating policies of reservoir system have significant effects on reliabilities of system. In multiple-reservoir operations, release policies improve reliabilities of the system. Limited reservoir storage reallocations from flood control to water supply uses did not have much effect on the system reliabilities. However, the proposed salt control dams had significantly improved the reliabilities, particularly, with stringent maximum allowable salt concentrations.