Balancing sediment connectivity and energy production via optimized reservoir sediment management strategies

Abstract

<p>Dam development projects cumulatively alter natural river connectivity, e.g., for fish and sediment, resulting in cumulative negative externalities across different spatio-temporal scales. Strategic siting of dams combined with dam-scale management, e.g., drawdown flushing for sediment passage, can help mitigate these impacts.</p><p>Designing optimal reservoir management strategies for multiple dams (i.e., dam portfolios), which account for economic objectives as well as sediment connectivity, is rarely done due to the lack of specifically designed modelling tools to properly quantify the hydro-morphological response of river systems to water and sediment management schemes. Models designed for this purpose must retain both a basin-scale perspective (to capture cumulative impacts of multiple dams) and a dynamic time representation (to capture dynamic reservoir operations).</p><p>This work presents a novel approach to reduce trade-offs between hydropower and sediment through integrating both optimal site selection and optimized joint operation of selected dam sites for sediment flushing. To estimate basin-wide sediment delivery and transport and quantify the effect of reservoirs on it, the study uses a new version of the D-CASCADE model, a process-based basin-scale dynamic sediment transport model.</p><p>The study focuses on the 3S river system, a data-scarce tributary of the Mekong river, where major dam development is ongoing. First, (1) the D-CASCADE model is set up and compared to available evidence of grain sizes and transport rates in the network. Then, (2) the effect of reservoir management is explored for different, pre-defined dam development portfolios focusing on downstream reservoirs, assessing daily sediment transport and delivery. Reservoirs features (i.e., volume, energy generation, and sediment storage) are dynamically simulated via integrated modelling add-ons. Finally, (3) sediment management (through drawdown flushing) is optimized by including parameters specific to the timing, frequency, and design of drawdown flushing into the operation rules.</p><p>Modelled network sediment yields matching field data measurements are identified and used as a baseline scenario to which to compare dam impacts on sediment delivery. Without sediment management, the model estimates a reduction in network sediment yield to the Mekong river of 32%-57%, depending on the dam portfolio. Sediment management portfolios showcase how reservoir sedimentation and downstream sediment starvation can be mitigated via well-designed flushing operations, albeit at a non-indifferent loss in energy production.</p>