Sediment transport, turbidity, and dissolved oxygen responses to annual streambed drawdowns for downstream fish passage in a flood control reservoir

Abstract

Sediment transport, turbidity, and dissolved oxygen were evaluated during six consecutive water years (2013–2018) of drawdowns of a flood control reservoir in the upper Willamette Valley, Oregon, USA. The drawdowns were conducted to allow volitional passage of endangered juvenile chinook salmon through the dam's regulating outlets by lowering the reservoir elevation to a point where the historical streambed was exposed and transported water and sediment through the reservoir dam. Sediment loads during the drawdown were highest in the first year of monitoring, with a computed value of 40,200 metric tons over a 5-day drawdown, followed by 5 years of lower sediment loads and lower sediment transport rates, suggesting that much of the stored sediment within the reservoir thalweg was transported out of the reservoir in the early years of the consecutive drawdowns. Suspended sediment concentrations (SSC) computed using turbidity and streamflow data resulted in maximum SSC at the onset of the drawdowns, with the highest computed values occurring during the water year 2017 drawdown at 17,500 mg/L (turbidity = 2,990 FNU), and average drawdown SSC values ranging from 654 to 3,950 mg/L for the six years of monitoring. Computed SSC were on the lower range of concentrations that could be harmful to out-migrating juvenile salmon published in other studies. High amounts of particulate organic matter and sand-sized material in drawdown SSC samples affected relations between turbidity and SSC, requiring the use of multiple surrogate regression models over short time frames. Dissolved oxygen minimum values were recorded in two of the monitoring years, with a minimum value of 0.71 and 3.4 mg/L recorded at the onset of the drawdowns in water years 2016 and 2018, respectively. Dissolved oxygen values below 4 mg/L lasted for 1 h, suggesting a rapidly expressed chemical oxygen demand. The response of suspended sediment loads and SSC highlight the site-specific nature of reservoir drawdowns, and the need for evaluation of expected sediment responses for drawdowns being considered at other locations.