Embracing Uncertainty as the New Norm: A Risk-Based Portfolio Approach for Urban Water Investment Planning

Providing secure and reliable water supply service to major urban areas has become a considerable challenge in recent years on a global basis. Rapid population growth, urbanization and development needs put enormous pressure on water resource managers to satisfy the ever-growing demand. Climate change, in addition to the inherent variability in hydrological cycles, adds another layer of deep uncertainty to forecast surface water availability. Many major cities have observed declining reservoir storages during unprecedented droughts. The once-reliable reservoir storage systems can no longer serve its purpose. During extended period of water shortages, urban residents and businesses suffered from mandatory water restrictions, causing large economic and social welfare losses. Facing these challenges, water utilities and governments make large investments in supply augmentation infrastructure, which have long-term consequences that can shape development for decades. However, the increasing complexity of uncertainty suggests that the ability to predict the future is limited; hence, there is a need to shift from the conventional "predict-then-act" planning paradigm. This thesis presents an alternative framework to urban water investment planning, using a portfolio approach.;A generalized risk-based framework for urban water supply-demand planning is proposed, and it is applied to Melbourne, Australia, to demonstrate its utility and usefulness. First of all, water shortage risk is clearly defined in two terms--frequency and severity of water shortages--of a defined planning horizon. Supply-side uncertainty is quantified based on probability distributions of precipitation and runoff to reservoirs. Demand-side uncertainty is modeled by scenarios with different combinations of population growth rate and per capita water usage. Next, the thesis presents an investment decision-making tool to identify cost-effective supply-demand portfolios that minimize water shortage severity while achieving a target level of reliable service. In addition to find the optimal portfolio composition, the model presents sequences of investments, indicating timing of implementation of each chosen measure. Using mixed integer programming, the decision-making tool yields Pareto efficient frontiers for different demand scenarios. The Pareto frontier exhibits trade-offs between cost of a water supply-demand strategy and water shortage risks facing a society in the long run. The trade-offs provide analytical insights on risk attitude towards water supply services, namely (i) what is the acceptable level of water shortage risk for a society, and (ii) how much are customers willing to pay to avoid such a risk. The results indicate that a portfolio which diversity risk of individual supply augmentation and conservation measures is robust when confronting a wide range of plausible climate and demand growth scenarios. Finally, recognizing important roles played by society and government in water-related investment decision-making process, the thesis discusses institutional barriers in adopting and implementing the proposed risk-based framework in practice.;This thesis presents an alternative framework to quantitatively integrate risk in urban water resources management. Under this framework, the portfolio approach is an analytical tool for decision-makers to prioritize investments in supply augmentation infrastructure and implementation of demand management programs. It is the hope of the author that this work provides new insights and necessary tools to water sector professionals in urban water investment planning. The use of risk-based framework and portfolio approach is not limited to any specific city and could find many applications in urban areas where water scarcity and climate risk are pressing issues.