The increasing popularity of low-impact certified hydropower has yielded a growing number of opportunities to examine the effects of turbine operations on fish survival, independent of other causes of mortality associated with downstream fish passage.
This is valuable information for developers and resource managers discussing the costs and benefits of proposed hydropower projects. A study was performed to assess fish passage survival before and after hydropower turbine installation at a water storage project.
Rimrock Lake, an irrigation water storage reservoir, is home to one of the largest and most robust populations of bull trout in the state of Washington (Kalin et al. 2002), a species listed as threatened under the U.S. Endangered Species Act (63 FR 31647). The coldwater tributaries that empty into Rimrock Lake, namely Clear Creek, Indian Creek, and the north and south forks of the Tieton River, are ideal for bull trout spawning and rearing. Subadult and adult bull trout find plentiful forage in the reservoir where there are large populations of kokanee salmon and other abundant prey species. Population estimates based on mark-recapture studies conducted by James (2002a) indicate that bull trout abundance in the reservoir varies annually between 1,024 and 3,392 mature adult bull trout with a mean estimate of 2,220.
High mortality rates
Tieton Dam, which forms Rimrock Reservoir, was built along with several other projects in the upper Yakima and Naches Rivers to store and supply water to downstream irrigators in the Yakima Valley—one of the most productive agricultural areas in the Pacific Northwest. The dam currently has no upstream fish passage, but fish from the reservoir are capable of passing downstream through the project via the intake tower near the center of reservoir. Prior to 2006, fish entrained at the intake were siphoned into a large pipe and exited the project through one of the two jet valves at the base of the dam.
Mortality estimates for fish passing through the project via the jet valves were quite high—81 percent in 2001 (James 2002b), 84 percent in 2002 and 80 percent in 2003 (E. Best pers. comm.). Given the condition of fish captured during previous entrainment studies (James 2002b, Hiebert et al. 2003, Hiebert et al. 2004) and the engineering of the project, researchers predicted that the primary source of mortality was the sudden pressure change, up to 5 atmospheres, and/or high velocities and associated sheer stress experienced by the fish as they rapidly traveled from the intake tower to the stilling basin.
In the late 1990s, a development company began assessing the feasibility of installing and operating a small, 13.6-MW run-of-the-river hydropower facility at the base of Tieton Dam. This eventually led to a formal Biological Assessment (BA) examining the effects of the proposed Tieton Hydroelectric Project on bull trout, which was completed in July of 2002. The central question was whether or not hydropower turbine operation would increase rates of entrainment and subsequent mortality of bull trout compared to existing project operations. The BA concluded that operation of the turbines would likely reduce mortality of entrained fish compared to existing operations, which channeled all flow through the high velocity jet valves. The number of entrained bull trout was not expected to be affected by operation of the hydropower facility because reservoir management and velocities near the intake structure would not change due to turbine operation. Turbines would be operated opportunistically when sufficient flow was released from the dam.
During development of the BA, researchers estimated that mortality of entrained fish passing through the project would decrease by more than 50 percent as a result of turbine installation and operation. This prediction was predicated on findings from survival studies conducted on small hydropower projects, which document substantially lower mortality rates compared to estimates of mortality for fish passing through the Tieton Dam jet valves. During high flow periods, a significant portion of water would still be discharged through the jet valves because the maximum flow capacity for the proposed Francis turbines was roughly 1,200 cubic feet per second (Kalin et. al. 2002). Therefore, the survival benefit associated with hydropower operations would only be realized by fish passing through the turbines; fish traveling through the jet valves would likely continue to survive at the lower rates observed by previous studies (~20%).
In agreement with the BA, the Biological Opinion (BO) prepared by the U.S. Fish and Wildlife Service concluded that operation of the Tieton Hydroelectric Project was not likely to jeopardize the persistence of bull trout, nor would it result in the destruction or adverse modification of critical habitat. However, one of the conditions of the BO was that entrainment mortality be studied following completion of the hydropower project (Lewis 2002).
In 2006, Tieton Hydropower LLC installed twin 6.8 MW Francis turbines on Tieton Dam. Availability of data from previous entrainment studies at the site provided a unique opportunity to compare current mortality rates to those observed prior to the installation of the turbines. The purpose of our investigation was to evaluate whether hydropower operations have increased mortality of fish entrained at Tieton Dam relative to pre-hydropower conditions. Present day operations allow a portion of flow and entrained fish to pass through two hydropower turbines. Water from the turbines is discharged into the stilling basin several meters below the surface. We hypothesized that the expected reduced mortality of fish passing through the turbines relative to the jet valves would improve overall survival through the project. To test our hypothesis, we collected a representative sample of fish entrained by the project in 2009 and compared our estimate of mortality to levels observed prior to hydropower development.
The Tieton River is a tributary to the Naches River located in central Washington. Both the Tieton and Naches Rivers are part of the Yakima River watershed, which drains the eastern slopes of the Cascade mountain range. The Yakima River flows southward through the arid Yakima Valley to eventually join the Columbia River 541 km upstream of the Pacific Ocean. Precipitation varies from approximately 128 inches along the crest of the Cascades to less than 8 inches on the eastern valley floor. Much of the precipitation falls as snow in the upper basin, and flows generally peak as snowmelt in the spring. Tieton Dam, 40 miles west of the city of Yakima, has an annual discharge of approximately 350,000 acre feet. Flow releases peak in the spring and fall, concurrent with spring runoff and the fall irrigation season.
From Aug. 28, 2009, to Oct. 11, 2009, an eight-foot rotary screw trap was operated below Tieton Dam. The trap was secured to the access road bridge about 200 meters downstream of the dam. Metal cables were used to position the trap such that it was fishing just downstream of the bridge in the primary stream channel. When operating, the trap was checked daily by a two-person crew. During maintenance or inoperative periods, the drum was lifted into the non-fishing position. During trap checks, debris build-up such as algae or plant material was cleaned off the trap and the trap was examined for proper mechanical operation. Fish were then removed from the livebox by dip net and placed into five gallon buckets. In preparation for handling, live fish were anesthetized with clove oil diluted in 95 percent alcohol. Each fish was identified to species and rated for condition factor. Fish condition was a qualitative assessment of external injuries, such as descaling, cuts, or missing fins. Fish were rated from I to IV as follows:
I. >75% scale loss, moderate bleeding, fin damage severe, missing head, eyeballs, etc.
II. 75 to 50% scale loss, noticeable bleeding, fin damage
III. 49 to 25% scale loss, no bleeding, little or no fin damage
IV. <25% scale loss, no bleeding, no fin damage
After handling, live fish were allowed to recover in a bucket of fresh water. Once conscious, the fish were released downstream of the trap.
After completion of field sampling, catch data was summarized by species for each day during the study period from Aug. 28, 2009, through Oct. 11, 2009.The proportion of total catch represented by bull trout in the sample was very low, so kokanee were chosen as a proxy to estimate bull trout entrainment mortality rates. This allowed for direct comparison to mortality estimates from 2001-2003 that were also calculated using kokanee. Given the size similarities between kokanee and juvenile bull trout, the life-stage most susceptible to entrainment, it was expected that survival rates experienced by bull trout would be very similar to rates observed for entrained kokanee. We examined the length distribution of kokanee to confirm the size range of fish susceptible to entrainment. Observed mortality rates for kokanee were calculated and compared to estimates of mortality derived from the previous studies using a Chi-square test. Several environmental variables including, pressure differential, rate of flow through the jets and turbines and forebay elevation, were collected during the sampling period. Previous studies identified a relationship between these variables and catch and mortality rates. We duplicated those analyses to see if the same relationships were present during the 2009 sampling period.
Catch data and species composition from rotary screw trap sampling in 2009 is summarized in. A total of 1,065 kokanee were caught in the screw trap during the study from late August to mid October. Of those, 478 fish were dead yielding an observed mortality rate of 45 percent. This mortality rate is significantly lower than the estimates derived before the installation of the Francis turbines, 80 percent to 84 percent, when fish were expelled from the high velocity jet valves exclusively, (chi-square test, p-value < 0.0001).
Contrary to previous studies, the majority of fish collected in 2009 had minimal exterior injuries and, therefore, received
high scores for condition factor. This was the case for both live and dead fish, most of which had very little descaling and no visible injuries. Of the fish collected, about 95 percent received the highest condition index score, indicating little or no signs of external injury.
The length frequency distribution of kokanee caught during screw trap sampling is shown in. Fish caught in the screw trap ranged from 20 mm to 275 mm with a mean of 133 mm. The distribution is bimodal, providing evidence that there are two distinct age classes represented in the sample. The length distribution observed during this study was similar to those observed in previous studies conducted at the same site in 2001-2003 (James 2002b, Hiebert 2003, Hiebert 2004).
Environmental variables were examined to determine their effect on entrainment and mortality rates for kokanee. Similar to James (2002b), we found a strong relationship between change in pressure and the proportion of dead fish collected in the samples, (p < 0.0001, R2 = 0.55). However, total catch was not linearly related to total flow as had been observed during previous fyke net studies. The amount of flow passing through the turbines and jet valves were also independently analyzed and neither was found to be a reliable predictor of mortality. This result is consistent with findings from screw trap sampling efforts conducted downstream of other water storage projects with similar outlet works (Symbiotics LLC 2009).
The increasing popularity of low-impact certified hydropower has launched a new generation of small, run-of-the-river projects integrated into existing water storage and flood control dams. Recent hydropower turbine installation at Tieton Dam provided a unique opportunity to study changes in downstream fish passage mortality rates associated with hydropower operations at an irrigation water storage facility. We compared fish entrainment mortality estimated after hydropower installation to estimates compiled prior to hydropower development. Entrainment of the key species of interest, bull trout, was rare; thus, it was necessary to use abundant kokanee salmon as a surrogate for estimating mortality rates for entrained bull trout.
Screw trap sampling below Tieton Dam yielded a total mortality estimate of 45 percent for kokanee entrained between Aug. 28, 2009, and October 11, 2009, nearly a 50 percent reduction in total mortality compared to previous estimates from 2001-2003 (James 2002b, E. Best pers. comm). Environmental data collected at the dam revealed that an increase in pressure differential is accompanied by an increase in water velocity at the jet valves. Both pressure differential and sheer stress are probable causes of mortality. In 2009, we observed a mean change in pressure of 65 psi which is greater than the mean value from 2001, 56 psi. The higher pressure differential in 2009 increased our confidence in the assertion that entrainment mortality declined significantly after hydropower development.
In 2001–2003, prior to turbine installation, fyke net sampling below the Tieton Dam revealed rates of kokanee mortality as high as 84 percent for fish entrained at the project during the high flow release period(James 2002b, Eric Best pers. comm). Researchers hypothesized that the primary source of mortality was the sudden pressure change and high velocities experienced by the fish as they traveled from the intake tower to the stilling basin. Kalin et al. (2002) predicted that the proposed installation of two Francis turbines on the outlet of the dam would provide an alternate route of passage for entrained fish and reduce mortality by as much as 45 percent for fish passing through the turbines. This prediction was based upon findings from the Electric Power Research Institute (1987), which reviewed 64 studies and concluded that peripheral runner velocity was the best predictor of turbine-induced fish mortality, particularly for Francis turbines. Furthermore, Hardin (2001) found a significant relationship (p = 0.0014, R2 = 0.58) between peripheral runner velocity and fish mortality when he examined 14 hydropower units operating Francis turbines similar to those installed at Tieton Dam. The relationship between peripheral runner velocity and predicted mortality is described as,
Percent Mortality = 0.4965 * (Peripheral Runner Velocity) - 17.919.
The peripheral runner velocity of the Tieton hydropower turbines is 93 feet per second, which yields an expected mortality estimate of 28 percent for fish passing through the turbines.
Similarities between mortality changes predicted by Kalin et al. (2002) and those we observed led us to believe that the significant reduction in entrainment mortality is likely a result of the installation and/or operation of the two Francis turbines. Most fisheries studies concerning hydropower facilities focus on assessing fish passage mortality increases resulting from turbine operation. Our study suggests that in some cases hydropower development may significantly reduce fish passage mortality relative to existing project operations.
Our observation of a 45 percent reduction in total mortality from 2001 to 2009 affirms the expected survival increase for fish passing Tieton Dam via the hydropower turbines, but it is also important to acknowledge the possibility that alterations to the outlet works may have improved survival through the jet valves as well. Project outflow is currently being routed through four outlets at the base of the dam—two jet valves and two turbines. Prior to hydropower installation, the entire volume of water was forced through the two jet valves. We expect that the sheer stress experienced by fish passing through the jet valves could have been reduced when a large portion of outflow was allowed to exit the dam through the turbines. However, this is somewhat speculative since water velocities at the jet valves were not directly measured before and after turbine installation.
Our findings suggest that modifications to the dam’s outlet works and operation of the Francis turbines did not increase fish entrainment and reduced the risk of mortality for bull trout passing through Tieton Dam, thereby providing a fisheries benefit. This observation suggests that typical assumptions about hydropower effects on fish should not be applied in all cases. Moreover, it appears that existing water storage projects warrant careful analysis to identify opportunities to reduce fish entrainment mortality through provision of alternate routes of dam passage, and hydropower development may provide a viable means of achieving this objective while simultaneously producing a sustainable source of energy.