Detroit Dam Structural Scenarios


Structural scenarios using the Detroit Lake model were limited only by the three possible types of outlets that are available in the USGS-coded CE-QUAL-W2 v3.12 blending routine: fixed-elevation, floating, or sliding-gate (variable-elevation). Sliding-gate outlets have user-specified vertical limits in the depth of the lake, whereas floating outlets are located at a user-defined depth below the water surface, and have a lower vertical limit. Similarly, sliding-gate outlets are positioned at a user-defined depth below the lake surface when the blending routine calls for such an outlet to be positioned at or near the lake surface. For this study, a lower vertical limit of 1,340 ft (the elevation of the upper ROs) and a depth of 6.6 ft (2 m) below the lake surface were specified for all floating and sliding-gate outlets. Four possible combinations of fixed-elevation, floating, and sliding-gate outlets, as well as the existing fixed-elevation outlets, were used in separate groups of structural scenarios in this study (table 6). 


Single Sliding-Gate Structure


Replacing all outlets at Detroit Dam with a single sliding-gate assembly simplified the modeled operations greatly. No minimum power generation constraints were set because presumably any or all of the flow through the new outlet could be routed to the hydropower plant. The minimum and maximum outflow rates of the base operations still applied. This structural scenario was named slider1340 because the lower elevation limit was set at 1,340 ft, the elevation of the upper ROs.


Scenarios in which a single sliding-gate outlet was used led to modeled outflow temperatures that generally varied more on a daily basis compared to scenarios using more than one outlet (fig. 20). This tendency was especially evident in autumn. The large variation in release temperatures is a result of the sliding-gate outlet being positioned at a depth that often was located in or near the middle of the thermocline, such that any seiching of the lake caused the thermocline to move up and down over the course of the day and thereby change the temperature of the water captured by the outlet. The model scenario was configured so that the elevation of the sliding-gate outlet was adjusted by the model only once per day (at 0500 hours), in order to minimize demands on dam operators; similar criteria were used for other scenarios such that gate adjustments were generally only performed once a day. Despite the larger daily variations in release temperatures, use of a single sliding-gate outlet in these slider1340 scenarios generally allowed downstream temperature targets to be met, with the exception of occasional spikes in autumn and general exceedances during December (fig. 20).


Floating and Fixed-Elevation Gates


Structural scenarios in which a fixed-elevation outlet and a floating outlet were used in combination (pp-float and uro-float) led to modeled release temperatures that were similar to results from scenarios with a single sliding outlet (slider1340), but the former generally displayed less daily variation than the latter. The pp-float and uro-float scenarios depict the existing power penstocks and upper RO gates, respectively, as fixed-elevation outlets used in combination with a new floating outlet. 


Floating Outlet with Power Penstocks


The pp-float structural scenarios specify one hypothetical floating outlet 6.6 ft (2 m) below the lake surface as well as a fixed outlet at the elevation of the existing power penstocks (1,402.9 ft centerline elevation). A series of operational scenarios were modeled in combination with these outlets to determine the effect of different minimum flow requirements to the power penstocks and the floating outlet. Results showed that as the minimum of the total outflow directed to the fixed-elevation (power) outlet was increased, the release temperatures generally decreased in spring and increased in the autumn. This effect is visible in figures 21, 22, 23, and E5 by comparing the discharge rate from the fixed outlet and the resulting outflow temperature in each scenario.


No Minimum Flow through Power Penstocks or Floating Outlet


With no minimum flow directed to either outlet, the model was free to optimize the release temperatures based on the lake temperatures near each outlet. As a result, the release temperatures generally met the max temperature target in scenarios c10 and n10. The hot/dry scenario (h10), however, resulted in an undesirable peak outflow temperature of about 54°F in autumn (fig. 21). 


Twenty Percent Minimum Flow through Power Penstocks


As the minimum outflow to the power penstocks is increased from 0 to 20 percent (scenarios c12, n12, and h12 in table 7), the amount of warm surface water that can be released via the floating outlet is decreased, resulting in cooler releases in June and July and warmer releases during autumn (fig. 22). The warmer releases in autumn are a direct result of releasing more of the deeper, cooler water in midsummer, which decreases the reserves of cool water at the level of the power penstocks and draws the thermocline down to deeper depths, thus pulling warmer water to the elevation of the power penstocks in autumn.


Minimum Flow of 400 Cubic Feet per Second through Floating Outlet 


The use of a floating outlet has two potential benefits. First, it allows continual access to warm water at the top of the lake in spring through autumn. Second, it can provide a means of collecting fish for downstream passage. Certain engineering design criteria and the use of the floating outlet for fish passage might require that the outlet be operated with a minimum flow rate. By placing a 400-ft³/s minimum outflow requirement on the floating outlet (fig. 23), the release temperatures are quite similar to those that result when no constraints are placed on either outlet (fig. 21). As with the pp-float scenario that required no minimum flows through either outlet, this scenario exceeds the target temperatures in autumn for the hot/dry environmental scenario, mainly because the lake level is low under those conditions and neither outlet is able to access water that is deep enough to still be cool at that time of year. 


Floating Outlet with Upper Regulating Outlets


The uro-float structural scenarios specify one hypothetical floating outlet 6.6 ft (2 m) below the lake surface as well as a fixed outlet at the elevation of the existing upper ROs (centerline elevation of 1,339.9 ft). These scenarios were developed under the assumption that outflow from upper ROs could be routed to the powerhouse at Detroit Dam for power production; therefore, the same sort of operational scenarios for a minimum amount of power generation were applied. 


Similar to results from the pp-float structural scenarios, as the outflow directed to the fixed-elevation outlet was increased, the outflow temperatures generally decreased in spring and increased in autumn. This trend can be seen in figures 24, 25, and 26 by comparing the discharge from the fixed-elevation outlet (labeled “Fixed out”) and the outflow temperature in each scenario. Greater releases of cool water from depth in midsummer generally diminish the probability of meeting June and July temperature targets and deplete the reservoir of cool water that is available for release in autumn.


Minimum Flow of 400 Cubic Feet per Second through Floating Outlet 


One potential floating withdrawal structure design for Detroit Dam would have the ability to convey fish downstream given a minimum flow requirement through the structure. When a 400-ft³/s minimum outflow requirement is placed on the floating outlet and no minimum flow is directed to the upper ROs, the result is that the vast majority of outflow is directed to the floating outlet during June–July to meet the max temperature target. This leads to cooler outflow temperatures in autumn, when scenarios c14, n14, and h14 generally met the temperature target (fig. 24). Exporting more heat from the lake surface in midsummer allows dam operators to reserve more of the cool water at depth for use in autumn. Decreased export of water from depth in midsummer also means that the thermocline is not drawn down as far, helping to retain access to cool water below the thermocline at the fixed-elevation outlet in autumn.


Twenty Percent Minimum Flow through Upper ROs


Requiring at least 20 percent of the total outflow to pass through the upper ROs in scenarios c15, n15, and h15 results in releases from the ROs that do not fall below about 250 ft³/s during summer months. Under the uro-float_20ppmin and uro-float_40ppmin scenarios, it was assumed that power production could be routed through the upper ROs. These scenarios generally result in outflow temperatures close to the max temperature target during autumn (fig. 25), showing that some minimum amount of power can be generated while still meeting downstream temperature targets. 


Forty Percent Minimum Flow through Upper ROs


By increasing the minimum outflow requirement on the upper RO gates to 40 percent, outflow from the upper ROs does not fall below about 500 ft³/s during summer months (fig. 26B). These scenarios generally result in outflow temperatures that exceed the max temperature target during November (fig. 26A). Clearly, as more water is drawn from below the thermocline in midsummer, less of the cool water below the thermocline is accessible in autumn.


Sliding and Floating Gates


Structural scenarios using a combination of a sliding-gate and a floating outlet were run to evaluate how access to both warm surface water and cool water at depth would allow downstream temperature targets to be met under a range of conditions. The sliding-gate outlet was assigned a lower vertical limit of 1,340 ft, which is similar to the elevation of the upper ROs. Similar to scenarios depicting a single sliding-gate outlet, these slider1340-float structural scenarios resulted in outflow temperatures near the max temperature target for most of the calendar year except for the month of December.


Fixed Flow of 400 Cubic Feet per Second through Floating Outlet


By simulating a constant flow of 400 ft³/s to the floating outlet, this scenario was designed to represent the potential effects of a hypothetical year-round lake-surface withdrawal structure that might also accommodate fish passage. Results from scenarios c19, n19, and h19 show that max temperature targets generally could be met throughout the year with this outlet configuration (fig. 27), although the temperature target was actually exceeded at times in August–September due to the large surface outflow. The USGS-coded CE-QUAL-W2 v3.12 blending routine does not explicitly solve for the mixed temperature between two outlets when a constant flow is designated to one outlet, so some of the exceedances in outflow temperature during late August and early October (fig. 20A) may be due to inconsistencies between the imposed temperature target and the blended outflow temperature calculated by the model. A modified blending subroutine could fix this problem, but the point is that this scenario can come close to meeting the temperature target most of the time. 


First posted October 30, 2012

Revised June 11, 2013