PREDICTION OF TOTAL DISSOLVED GAS CONCENTRATIONS DOWNSTREAM OF SPILLWAYS

摘要

Dam spillways are opened to allow juvenile salmon to migrate past the series of dams on the Snake and Columbia Rivers, USA. Unfortunately, spillway operation causes high aqueous concentrations of oxygen and nitrogen, or total dissolved gas that is harmful to salmon. The concentration of total dissolved gas (TDG) will vary depending on the dam's geometric configuration, as well as hydraulic and operating conditions. Field measurements are required at each structure to predict TDG concentrations for each set of operating conditions. Quality field measurements are difficult and expensive to obtain, so a computer model that could predict TDG concentrations for different conditions and configurations would save considerable cost and effort in gathering field data at each structure. Current predictive models are only applicable to the spillway and conditions under which the data were collected because the modeling of the physics governing gas transfer is incomplete within these models. A new predictive model was developed that focuses on the physical processes of gas transfer. The goal was to develop a more accurate and widely applied model that would not require extensive fieldwork at each dam. Model components are described, model predictions are compared to field data, and a sensitivity analysis gives insight into the most important parameters controlling gas transfer.