UC Santa Cruz

Appropriate water supply scenario modeling, that is, modeling techniques that capture the fullest range of dynamics observed in real hydrological systems, will be an essential tool in meeting 21st century water management challenges. This is especially salient in regions such as California, with large, water-dependent populations, agriculture and industry; complex water delivery systems; highly variable precipitation regimes; extended and pervasive droughts; and uncertainty in future water supply estimates associated with the potential effects of climate change. While scenario modeling is crucial, current methods often have limited incorporation of long-run forms of persistence, multi-year extremes, and a range of potential climate change effects. This dissertation addresses these applied research issues to improve historical water supply scenario modeling and highlight potential risks associated with climate change for inflows to the important Shasta/Trinity and Oroville Reservoirs of California. The goals of this dissertation were to: (1) identify and characterize long-run persistence in inflows; (2) incorporate this persistence, especially in the forms of hydrologic extremes, into water supply scenario modeling; (3) generate a range of potential hydrological futures under climate change for these inflows; and (4) examine supply reliability challenges and the role of scenario modeling in California water agencies. Key findings indicate that: (1) long-run persistence, related to large-scale climate oscillations such as the El Niño Southern Oscillation (ENSO) and the Pacific Decadal Oscillation (PDO), contribute to a significant portion of inflow variability; (2) use of an innovative hybrid Empirical Mode Decomposition (EMD)-Matalas modeling framework overcomes some of the limitations of traditional models to more accurately model modes of long-run persistence and multi-year extremes; (3) climate change presents potentially broad shifts and high uncertainty in future hydrological inflows where increased variance affects the frequency, duration, and intensity of floods and droughts; and (4) under this large uncertainty a water supply agency might wish to consider their risk profile and potential barriers to change.