Scientific Investigations Report 2007–5008
U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2007–5008
A CE-QUAL-W2 model of Detroit Lake was constructed and calibrated for calendar years 2002 and 2003, and for a period of heightened storm inflows between December 1, 2005, and February 1, 2006. The model effectively simulated the lake’s spatial and temporal patterns in hydrodynamics, water temperature, TDS, and two size groups of suspended sediment. Most sediment entered the reservoir during storm events. The lake is an efficient sediment trap, capturing more than 90 percent of the incoming suspended sediment in each of the modeled time periods. The model results indicate that a large percentage of suspended sand and silt was deposited in the upper reaches of the reservoir; this size of sediment typically only exited the reservoir in the outflow during large storm events. In contrast, clay-sized sediment stayed in suspension for longer periods of time, and composed most of the suspended sediment in the reservoir outflow. The source of the suspended sediment in the outflow came from all tributaries to Detroit Lake, but the largest contributions were from the two largest tributaries, the North Santiam and Breitenbush Rivers. A larger percentage of the suspended sand and silt in the outflow came from tributaries closer to the dam, because this size of suspended sediment settles rapidly.
Water temperature in Detroit Lake followed a seasonal pattern, with surface warming in spring, development of a thermocline in summer, and surface cooling and turnover in autumn. At present, the temperature of water released from Detroit Lake is cooler in summer and warmer in autumn compared to a more natural seasonal temperature pattern. Simulation of a hypothetical selective withdrawal device showed that with structural and operational adjustments, the outflow from Detroit Lake could approximate a more natural seasonal temperature pattern for most of the year, but also that the lake’s supply of cold water might be exhausted in autumn.
Modeling the spatial patterns of sediment deposition in the reservoir could be improved by including algorithms to simulate sediment scour and resuspension. This would allow a more detailed examination of the spatial patterns of sedimentation in the reaches near the major inflows that are affected by seasonal drawdown. The current Detroit Lake model also does not distinguish between inorganic and organic sediment. It would be useful to collect more data to determine the makeup of the inflow suspended material, as CE-QUAL-W2 is able to simulate degradation of particulate organic material. Simulation of the 2005–06 storms showed the importance of large storm events on sediment transport into and out of the reservoir, and the model could be used to further assess future or historic storm inflows and suspended sediment transport and fate. For example, the large 1996 flood event would be a good candidate for further use and testing of the model. The model also could be used to assess patterns of sediment deposition in the lake over long time periods.
The use of a selective withdrawal device could be modeled in more detail, examining how different combinations of outlets, various downstream temperature targets, changes in reservoir operation, and modified release schedules might affect downstream water temperatures. An existing North Santiam River CE-QUAL-W2 flow and water temperature model could be upgraded and connected to the Detroit Lake model in order to predict how these changes to Detroit Lake might affect water temperatures on the North Santiam and Santiam Rivers downstream to the confluence with the Willamette River. Extension of the Detroit Lake model to include the Big Cliff reregulating reservoir would be a logical component of that effort.
The Detroit Lake model could be expanded to include nutrients and algae dynamics. Blue-green algae blooms are a potential concern because of their associated toxins and their known occurrence in Detroit Lake, albeit at low levels. Significant blooms have occurred in other Cascade Range reservoirs, leading to health advisories and recommendations that visitors avoid contact with lake water. In order to simulate algae in the Detroit Lake model, additional data on nutrients, chlorophyll, and algae would be needed.
U.S.
Department of the Interior | U.S. Geological
Survey
Persistent URL: https://pubs.water.usgs.gov/sir20075008
Page Contact Information: Publications Team
Page Last Modified: Thursday, 01-Dec-2016 19:40:52 EST
