Cascading Model-Based Framework for the Sustainability Assessment of a Multipurpose Reservoir in a Changing Climate
Wenxin Xu,Jie Chen,Tianhua Su,Jong-Suk Kim,Lei Gu,Joo-Heon Lee
DOI: https://doi.org/10.1061/(ASCE)WR.1943-5452.0001501
IF: 3.457
2022-01-01
Journal of Water Resources Planning and Management
Abstract:Climate change impacts on hydrological processes can affect reservoir operational performance. Hence, the reservoir operation model, based on historical climate conditions, may not guarantee sustainable water resources management in the future. To enable stakeholders to design reliable adaptation strategies, this study aims to propose a cascading framework to quantify the impacts of climate change on the operational performance and sustainability of a multipurpose reservoir. The Danjiangkou Reservoir (DJKR), which serves as the water source for the middle route of the South-to-North Water Diversion Project in China, was selected as a case study. To achieve the aforementioned aims, bias-corrected simulations from 13 global climate models (GCMs) were first input into five hydrological models [i.e., one data-driven [deep belief network (DBN)], three conceptual [SIMHYD, HBV, and Xin'anjiang (XAJ)], and one physically-based [variable infiltration capacity (VIC)]. The simulated reservoir inflows were then fed into a 10-day reservoir simulation model where DJKR operation followed the designed operating rules to evaluate reservoir operational performance. Finally, a data envelopment analysis (DEA) model was proposed to assess reservoir sustainability under both historical (1976-2005) and future (2021-2050) climate conditions. The results show that the combination of the GCM ensembles and the SIMHYD, HBV, XAJ, and VIC models exhibit similar growth patterns in the reservoir inflow and operational benefits for the future period. However, the DBN model produces consistent decreases in most cases, which may be attributed to its inability to generate accurate estimates of extreme events. The results indicate that hydrological models may be extensively utilized in decision making with greater confidence, and the data-driven model should be interpreted with caution when used in hydrological climate change impact studies. The efficiency metrics suggest that decision makers should focus more on increasing operational benefits, which can subsequently enhance reservoir sustainability. Overall, the framework proposed in this study provides a foundation for evaluating the reservoir sustainability and adaptability to climate change from water managers' perspective.