Downstream Effects of Selected Scenarios 


To assess the potential downstream effects of operational and structural changes at Detroit Dam, the Big Cliff Reservoir and North Santiam-Santiam River models were used to run a subset of the Detroit Dam scenarios. Not all scenarios were run through the downstream models because the downstream patterns of temperature change are likely to be similar. The selected scenarios were chosen because of their potential for being adopted by USACE as they evaluate possible operational and/or structural changes to Detroit Dam.


Estimated Emergence Dates


The Accumulated Thermal Unit (ATU) is a “degree-day” type of calculation used to estimate the date on which spring Chinook salmon first emerge from their eggs (U.S. Army Corps of Engineers, 2012). The ATU calculation in this report is the cumulative sum of the daily average temperature (in degrees Fahrenheit) exceeding 32°F beginning at the typical peak of spring Chinook spawning on September 20. The estimated emergence day then is derived as the date when the ATU value reaches 1,750°F-day. These emergence day estimates are based on observed egg emergences at the Oregon Department of Fish and Wildlife Willamette Hatchery in Oakridge, Oregon, when the ATU value is 1,650–1,850°F-day (U.S. Army Corps of Engineers, 2012).


Simulated Detroit Dam release temperatures were used for this computation. Some model scenarios resulted in estimated emergence dates that were later than December 31. In these cases, the simulated Detroit Dam outflow temperature from each respective scenario during the previous January and February were used to complete the ATU calculation. Estimated emergence dates under each Detroit Lake model scenario are compared and sorted according to the hot/ dry environmental scenario results in table 8. The hot/dry scenarios display a somewhat worst-case potential effect that any Detroit Dam operational/structural scenario might have on downstream water temperatures during a low-flow year. 


Scenarios having only operational changes at Detroit Dam resulted in earlier emergence dates than did scenarios with structural changes or structural and operational changes combined, suggesting that some structural changes might be more beneficial for fish under hot/dry conditions. Operational scenarios with lower minimum power production requirements, which generally provided cooler release temperatures in autumn, resulted in later estimated emergence dates for each otherwise equivalent scenario. For example, scenario h2 (existing structures, noppmin operations, hot/dry conditions) resulted in an emergence date 5 days later than h1 (existing structures, base operations, hot/dry conditions) (table 8). 


Emergence date estimates for many of the operational scenarios (c1, n1, c2, n2, c5, n5, c6, n6, c7, n7) under cool/ wet and normal environmental scenarios do not follow the same pattern as that exhibited by the hot/dry environmental scenarios; instead, the dates are later than those from many of the structural scenarios. This result is primarily caused by outflow temperatures that were well below the temperature target during early to mid-October, a time when the water-surface elevation was below the spillway crest and releases were limited to cooler water from the upper ROs and the power penstocks. The ATU computation has the advantage of integrating all of these conditions into a single numeric value, but it does not convey the rate at which the emergence date is approached or any sequence of events that changes that rate during autumn. Those changes in ATU over time are illustrated in figure 28, showing that many of the scenarios are likely to have quite different effects on the rate of egg incubation during autumn months (fig. 28).


Downstream of Big Cliff Dam in the North Santiam River, estimated emergence dates were calculated from model results (scenarios h1, h8, h10, h17, and h19) at the location of the USGS gaging station at Mehama (site 14183000, RM 38.7, table 9). All emergence date estimates at Mehama range from 15 to 43 days later than those calculated at the outlet of Detroit Dam. The cooler water temperatures at that site are caused partly by cool inflows from the Little North Santiam River just upstream of Mehama in November–December.


North Santiam River Temperatures


To examine the potential downstream effects of select scenarios at Detroit Dam, outflows from the Detroit Lake model were routed through the downstream Big Cliff Reservoir and North Santiam/Santiam River models. For these simulations, the South Santiam River inflow and all other tributary inflows to the North Santiam River were taken from measured conditions for the environmental scenario of interest. A 7-day moving average of the daily maximum (7dADM) water temperature in each model segment of the North Santiam and Santiam River model was used to display an image of the modeled river temperatures throughout the calendar year of the hot/dry environmental scenario. 


Base Case


The downstream conditions resulting from current base operational rules in place at Detroit Dam under the hot/dry environmental forcing conditions show that downstream river temperatures generally increase from the dam release temperatures, except in November and December when river temperatures decrease (fig. 29). Notably, the loss of spillway control at Detroit Dam around August 1 and the change in the temperature target near October 1 are apparent in the figure at RM 59 and can be traced through the river model downstream to the confluence of the North and South Santiam Rivers. The relatively large temperature differences due to operational changes at RM 59 during these times diminish with downstream distance through the river model. The effects of operational changes at Detroit Dam are reduced farther downstream of the confluence of the North and South Santiam Rivers from August through October due to the warming effect of the river system downstream of Big Cliff Dam. In early November, the North Santiam and Santiam Rivers below Big Cliff Dam begin to have a downstream cooling effect.


Delayed Drawdown


By reducing the minimum outflow requirements from Detroit and Big Cliff Dams from September 1 to October 16 as in operational scenario delay_dd2, the Detroit Lake water level remained higher than in base scenarios (compare figs. 7 and 17). However, this decreased outflow from Detroit and Big Cliff Dams under the h8 scenario led to downstream warming in the North Santiam River model results that exceeded the base h1 scenario. Although the outflow temperatures from Big Cliff Dam in h8 were cooler than those in h1 during autumn, the reduced flow in the North Santiam River led to temperatures 2–5°C warmer than in the h1 scenario (fig. 30).


Floating and Fixed-Elevation Gates 


The simulation of a hypothetical floating gate combined with the existing power penstock outlets (fixed-elevation gates) at Detroit Dam, with no minimum power-generation requirement (scenario h10, table 7), allows the Detroit Lake model to expel warm water during summer beyond the day in which the lake elevation falls below the spillway outlets in the h1 scenario. During a hot and dry year, the lake level may be lower than normal and not allow the use of the spillway late in summer; using a floating outlet circumvents this problem. This scenario has the effect of rationing cool water until later in autumn and cooling the North Santiam River about 1°C compared to scenario h1 throughout October and the latter half of November (fig. 31). Although that figure shows substantial warming in August in scenario h10 relative to h1, that warming is somewhat artificial because the spillway could no longer be used in the h1 scenario and cooler-than-desired water was being released in that case.


A Single Sliding-Gate Structure


Additional flexibility and an ability to meet the cool temperature targets in autumn is realized in model scenario h17 (table 7), in which a sliding-gate outlet ranging from 2 m below the lake surface to the elevation of the upper ROs is simulated at Detroit Dam. Scenario h17 temperatures between Big Cliff Dam and the South Santiam River confluence were generally 0.5–2.5°C cooler than temperatures for scenario h1 during October and November (fig. 32). Again, the warming in August shown in figure 32 is somewhat artificial as a result of the loss of the use of the spillway in early August in scenario h1; scenario h17 does a better job of meeting the max temperature target at that time. Downstream of the South Santiam River confluence in the Santiam River, scenario h17 continued to have a cooling effect throughout autumn as water temperatures remained as much as 1°C cooler than the h1 scenario from mid-October to mid-December. 


Floating and Sliding-Gate Structures


Similar to the sliding-gate-only structural scenario h17, scenario h19 specified the use of the same sliding-gate outlet with a lower vertical limit of 1,340 ft, but added a floating outlet with a fixed outflow year-round of 400 ft³/s (table 7). This scenario resulted in autumn release temperatures that were cooler than h1 (0.5–11.5°C in October and 1.5–3.0°C in November) from RMs 59 to 40 (fig. 33). Downstream of the confluence of the Santiam River near RM 12, the temperature effects from h19 are similar to h17.


First posted October 30, 2012

Revised June 11, 2013