Development of a parametric simulation model for reservoir operation optimization

Reservoirs are surely one of the most important hydraulic structures for the management of integrated water resources. In the last 30 years, with the remarkable increases in greenhouse gas emissions, the impacts of climate change on the hydrological processes have become much more noticeable, and that situation has brought about struggles in the operation of the reservoirs. In this regard, it has become more important to prepare climate projections and to designate operational policies that will adapt to these dynamic factors to ensure the future sustainability of the water needs demanded from the reservoirs and to plan possible precautions as well. In this study, which was implemented over Çaglayan reservoir to address the grounds above, the output of GCM derived under different RCP scenarios were downscaled to monthly precipitation and temperature through an artificial neural network-based strategy. The systematic biases in meteorological projections involving the 2021-2050 period were then corrected with the quantile delta mapping algorithm. These bias-corrected outputs were transformed into inflow projections employing a lumped hydrological modelling procedure. Afterward, it was intended to manipulate a parameterization-simulation-optimization framework that can adapt to scenario-GCM variations derived. At this phase, the RLS volumes to be optimally released from the reservoirs were obtained by effectively integrating the hedging model termed HDG-2D with a modified differential evolutionary algorithm. Finally, The results reveal that the developed model offers lower vulnerability and higher sustainability conditions under climate change variants compared to the standard operating policy.