Thesis (Ph.D.)--RAND Graduate School, 2005. Includes bibliographical references.

Introduction -- California's Water Resources and Future Challenges -- Standard Decision Theory and Its Limitations for Long-term Water Resources Planning -- Scenarios of Future California Water Demand -- A New Analytic Method for Identifying Robust Policies -- Robust Water Management Strategies for Southern California -- Summary and Key Insights -- Appendix 1: Additional Water Demand Scenario Results. Ensuring sufficient, high-quality water supplies for California over the next several decades will be a great challenge for water resource managers. Choosing an appropriate management response using standard methods will be extremely difficult and contentious because the scope and magnitude of these impacts are highly uncertain and stakeholders have diverse views about desirable outcomes. This dissertation first documents the development and use of a model to generate quantitative scenarios of future water demand in California. It next describes a new analytic method for decision making under deep uncertainty called Robust Decision Making (RDM). To demonstrate how RDM can be a valuable analytic tool for California long-term water planning, the dissertation applies the methodology to a stylized representation of the water supply and demand management challenge facing Southern California.