AUTHOR=Martinsen Grith , Liu Suxia , Mo Xingguo , Davidsen Claus , Payet-Burin Raphaël , Bauer-Gottwein Peter TITLE=The Impact of Assuming Perfect Foresight in Hydroeconomic Analysis of Yellow River Diversions to the Hai River Basin, China: A Framework Combining Linear Programming and Model Predictive Control JOURNAL=Frontiers in Water VOLUME=3 YEAR=2021 URL=https://www.frontiersin.org/journals/water/articles/10.3389/frwa.2021.648934 DOI=10.3389/frwa.2021.648934 ISSN=2624-9375 ABSTRACT=

This study analyses the impact of assuming perfect foresight of future agro-hydrological events in hydroeconomic analysis of water infrastructure projects. The impact is evaluated based on the estimated monetary benefits of a proposed water infrastructure investment diverting Yellow River water to the Hai River basin in China, resulting in supply augmentation and improved water quality. The impact of foresight is quantified as the change in project benefits, evaluated with different assumed lengths of future foresight compared to a perfect foresight benchmark. A hydroeconomic optimization model formulated as a deterministic Linear Program, LP, is optimized to represent the perfect foresight benchmark. Imperfect foresight is modeled by wrapping the hydroeoconomic optimization model in a Model Predictive Control, MCP, framework. Using this LP-MPC framework, different lengths of foresight can be modeled by continuous re-optimizations with updated forecasts over a planning horizon. The framework is applied to the water-scarce and polluted Hai River basin in China, which is suffering from groundwater overdraft and is dominated by agricultural irrigation demands. The hydroeconomic optimization model describes the nine largest reservoirs in conjunctive use with the major groundwater aquifers. The water infrastructure project, allowing transfers of Yellow River water to the plain area of the Hai River basin, is evaluated under long-term sustainable groundwater abstraction constraints, and joint water allocation and water quality management. The value of foresight in agricultural water allocations is represented, using a model that links yield response to water allocations, accounting for delayed yields in agricultural irrigation. Estimated benefits of the proposed project evaluated with decreasing lengths of foresight and compared to the perfect foresight benchmark show that an assumption of perfect foresight underestimates the actual benefits of the water infrastructure investment in the irrigation intensive Hai River basin. This study demonstrates that it is important to evaluate the impact of assuming perfect foresight in any hydroeconomic analysis, to avoid misleading conclusions regarding the costs and benefits of planned projects.