<p> In regulated rivers, reservoir operation decisions largely determine downstream river temperature and flow. Computational methods can minimize the risk and uncertainty of making regrettable environmental release decisions and aid operations planning and performance prediction. Mathematical modeling in particular can optimize the timing and magnitude of reservoir release decisions for downstream benefit while accounting for seasonal uncertainty, water storage impact, and competing water demands. This dissertation uses optimization and modeling techniques, modifying traditional optimization modeling to include temporal correlation in outcome variables and incorporating long-term planning and risk management into prescribed reservoir operations. The proposed method is implemented in one case, a) with a state variable that tracks outcome benefits over time (fish population size) and, in another case, b) with a maximin stochastic dynamic program solution algorithm that maximizes net operational benefit and minimizes worst-case outcomes (for cold water habitat delivery). This method is particularly useful for environmental flow management, when the water quality and quantity of the river and reservoir in one time step affect the quantity and quality in the reservoir and the river for later periods. Better solutions with these methods internalize risk and hedge releases at the beginning of an operating season to maximize downstream benefit and reduce the probability of catastrophe for the season and future years. Maximizing the minimum cold-water habitat area over months of a season or multiple years, or maximizing a river indicator variable explicitly, could likely help, for example, maximize an out-migrating salmon smolt population downstream. The method is demonstrated with a case study optimizing environmental releases from Folsom Dam and another optimizing temperature management from Shasta Dam in northern California. These results inform general rules for environmental flow management and temperature management of reservoirs, with specific policy recommendations for both Folsom and Shasta reservoirs. In both cases, the added value from employing hedging rules help reservoir operations minimize the risk of environmental catastrophe and conserve storage both within an operating season and across years.</p><p>
| Identifer | oai:union.ndltd.org:PROQUEST/oai:pqdtoai.proquest.com:10973511 |
| Date | 16 March 2019 |
| Creators | Adams, Lauren |
| Publisher | University of California, Davis |
| Source Sets | ProQuest.com |
| Language | English |
| Detected Language | English |
| Type | thesis |
Page generated in 0.0025 seconds