Water scarcity is intensifying globally due to climate change, economic growth, and regulatory shifts. Many regions are investing heavily in new water infrastructure, such as reservoirs, canals, and groundwater recharge facilities, to improve water supply reliability and mitigate impacts on various sectors. However, in the U.S., federal and state funding for water systems has decreased, leading to a reliance on local funding sources, primarily municipal bonds. This necessitates careful balancing of the benefits of supply investments against long-term debt obligations, especially considering the financial vulnerability of drought-affected providers. Higher borrowing rates and water affordability challenges further complicate matters. Collaborative partnerships offer potential cost savings and improved access to capital, but they also introduce complexities due to the inherently uncertain nature of water supply systems and the need to consider the distributed benefits and costs among partners. Existing economic planning frameworks often rely on aggregated models and expected value benefit-cost analyses that don't adequately capture the local-scale trade-offs and uncertainties involved. This research addresses this gap by developing a novel approach for designing resilient and equitable water infrastructure partnerships, focusing on the San Joaquin Valley in California as a case study. This region faces severe challenges from climate change, groundwater overdraft, and the need for major infrastructure investments. The study aims to provide guidance to water providers on navigating the complex tradeoffs associated with regional infrastructure partnerships under uncertainty, helping to enhance the resilience and financial stability of water supply systems.

The literature review examines existing approaches to water infrastructure investment and partnership design. It highlights the limitations of traditional cost-benefit analyses and game-theoretic methods in handling the complexity and uncertainty inherent in water supply systems. The authors discuss the challenges of assessing the value of groundwater recharge facilities, which require detailed modeling of both short-term and long-term dynamics, as well as the impact of interconnected water systems and complex institutional frameworks. Existing models often lack the necessary spatiotemporal resolution to capture the heterogeneity of water supply benefits and financial risks for individual partners within collaborative investments. The review concludes that there's a critical need for new frameworks that incorporate detailed modeling and robust optimization techniques to design partnerships that are both resilient and equitable.

This study uses a combination of detailed ensemble modeling and multiobjective intelligent search to analyze infrastructure partnership design. The researchers leverage the California Food-Energy-Water System (CALFEWS) model, a high-fidelity water resource system model capable of simulating daily reservoir operations, water rights, and conjunctive surface and groundwater management at the level of individual water providers. This allows for a fine-grained assessment of water supply benefits and financial risks for each partner. The model is applied to the Tulare Lake Basin in California's San Joaquin Valley, focusing on two key infrastructure initiatives: Friant-Kern Canal rehabilitation and the development of a new groundwater bank. The study considers multiple alternative infrastructure partnerships using an ensemble of 30-year daily time-step hydrologic sequences generated from a synthetic streamflow generator, accounting for the hydroclimatic variability of the region. A multiobjective intelligent search algorithm (Borg MultiObjective Evolutionary Algorithm – MOEA) is employed to explore approximately 300,000 candidate partnerships, identifying the optimal trade-offs between four key metrics: partnership size, water supply benefits for partners, water supply impacts on non-partners, and financial risk (measured by the cost of gains for the worst-off partner). The partnerships are evaluated based on the mean performance across the ensemble of hydrologic scenarios and the 90th percentile to assess financial risk. The researchers also compare their findings to the status quo planning process, represented by the Friant Contractors' current canal rehabilitation partnership.

The study's key findings reveal substantial tradeoffs in water infrastructure partnership design. The optimal partnerships, identified by the MOEA, typically invest in both canal expansion and groundwater banking, suggesting synergy between these projects. Partnership size is found to be highly correlated with financial risk, with larger partnerships tending to have higher costs for the worst-off partners. A significant trade-off exists between the water supply benefits for partners and the impacts on non-partners, implying potential for conflicts in water allocation. The analysis shows that even under optimistic hydroclimatic conditions, significant uncertainty remains, with substantial variation in water supply benefits and financial risks across different hydrologic scenarios and partners. The study demonstrates the advantages of their framework by comparing it to the existing Friant-Kern Canal rehabilitation partnership (the "Status Quo Partnership"). The analysis reveals that alternative partnerships found by the multi-objective search yield considerably better outcomes, showcasing significant "regret" associated with traditional, ad-hoc planning practices. One example partnership achieves 58% higher water supply gains while also significantly lowering extreme cost burdens for investing water providers. The heterogeneity in partner-level benefits and risks emphasizes the importance of detailed, localized modeling. The study found that the largest partnerships tend to create the greatest financial risks, suggesting a limit to the benefits of large-scale collaboration. The study also identified a strong trade-off between water supply benefits for partners and non-partners, indicating that gains for some may come at the expense of others. This raises concerns about stakeholder engagement and potential conflicts during the planning process.

The findings highlight the importance of a comprehensive approach to water infrastructure investment planning that explicitly considers financial risk, hydroclimatic uncertainty, and stakeholder equity. The significant performance trade-offs identified emphasize the limitations of traditional planning practices, which often fail to capture the nuanced dynamics of complex water systems. The study's results demonstrate the value of combining detailed water supply modeling with multiobjective optimization techniques to design more resilient and equitable partnerships. The research underscores the need for state and national agencies to provide more guidance on managing external stakeholder impacts. Furthermore, the findings suggest that integrated planning across multiple infrastructure projects may yield greater benefits. This study emphasizes the need for more comprehensive planning that considers the complex interplay between hydrological, infrastructural, and institutional factors.

This study demonstrates the critical need for a more sophisticated approach to water infrastructure partnership design. The severe trade-offs identified between supply, risk, and equity demand a departure from traditional ad hoc planning and the adoption of methodologies such as detailed ensemble modeling and multiobjective optimization. This framework enables the identification of partnerships that offer substantial improvements in water supply while mitigating financial risk and promoting equity among stakeholders. Future work should investigate the integration of these computational tools with stakeholder-based planning, explore contractual innovations to manage risk, and further examine the complexities of equity and power dynamics in water resource management.

While the study utilizes a high-fidelity water resource model and a robust optimization framework, some limitations exist. The optimistic assumptions regarding hydroclimatic conditions neglect potential non-stationary climate change effects. The analysis also focuses primarily on financial risk and water supply, while other important factors, such as social and environmental equity, might need more explicit consideration. The computational demands of the multiobjective search might pose a barrier for some water providers, highlighting the need for greater capacity building and technology transfer.