The immediate and cumulative impacts of irrigation withdrawals, loss of mainstem littoral habitat through shoreline development, dredging, navigation, and point and nonpoint source pollution have highly impacted anadromous fish stocks as they spawn, rear, and migrate through the mainstem Snake and Columbia rivers. While these activities have had profound impacts, they are largely unquantifiable. Through many years of extensive research, the impacts of construction and operation of mainstem and tributary hydroelectric projects have been quantified. The impacts of hydroelectric projects have been particularly responsible for anadromous fish declines through direct kills, selection against critical life history diversity, and profound changes to the ecosystem and trophic structures in which salmonids evolved (NWPPC 1986; Mundie 1991; Diamond and Pribble 1979; Petts 1980).
The Columbia River basin is probably the most dammed river system in the world. Since the construction of the first dams on the Willamette and Spokane rivers in the late 1800’s, a total of 136 dams for hydropower and other purposes have been built in the basin (NWPPC 1986). The impounded portions of these rivers have undergone significant environmental change from their lotic, or free-flowing, ecology to biological and physical conditions associated with standing bodies of water. The migrations of juvenile salmon through the impounded mainstem sections of the Columbia and Snake are significantly affected as a result of the dams. For instance, juvenile salmon originating in the Lochsa River in Nez Perce Tribe’s ceded area in central Idaho must traverse eight hydroelectric projects and approximately 300 miles of impounded river before reaching the Pacific Ocean.
Estimates of cumulative mortality from the effects of hydropower development and operation range from 35% to 96% in the juvenile life stage (NMFS 1995). Reductions in smolt-to-adult survival have coincided with increased numbers of dams, turbines, increased storage capacity, decreased spill, and decreased flow (Raymond 1988; Petrosky 1991). Attempts to isolate and quantify the magnitude of mortality resulting from various components of the hydrosystem are difficult, because the sources of mortality do not operate independently from one another on affected salmon populations. None-theless, numerous studies have addressed major impacts–e.g., water flow, turbines, water quality, spill, transportation, and structural barriers–regarding juvenile passage through the hydroelectric system.
Selective mortality to particular life histories from hydrosystem impacts may be responsible for large reductions in overall population diversity and viability (Lichatowich and Mobrand 1995). Schluchter and Lichatowich (1977) showed that diversity in life history types influenced juvenile growth rates, which in turn increased the variability at age of maturity, and that the existence of variable life history types lend stability to a stock. Different juvenile life histories have different migration times and rearing areas, thus there is a reduced probability of a single environmental disturbance impacting all life histories. They suggested that the annual abundance of adults in a stock containing multiple age classes would be stabilized because impacts to the variability of successes among broods would be distributed over several age classes in future adult populations.
Water Flow: Columbia and Snake river spring and summer flows have been substantially reduced as a result of hydropower, flood control, navigation, and irrigation development (Figure 5B.3). Reduced flows have changed the geometry of the river channel and water quality, and have increased the travel time of juvenile salmon migrating to the ocean (CBFWA 1991; Cada et al. 1993). Increased travel time reduces juvenile salmon survival as a result of: residualism in reservoirs; reduced ability of juvenile migrants to tolerate saltwater; delayed entry into the estuary and ocean past the period when conditions are most favorable for survival; increased exposure to predators; and exposure to increased water temperatures (Jaske and Goebel 1967; Diamond and Pribble 1978; Li et al. 1987; Vigg 1988; CBFWA 1991; Karr et al. 1992).
Turbines: Raymond (1979, 1988) and Anderson (1988) note that anadromous salmonid stocks in the Columbia River Basin are declining as a result of numerous factors, including installation and operation of turbines at mainstem Snake and Columbia river dams. Downstream migrating juveniles incur direct and indirect mortality when they pass through mainstem Kaplan turbines at hydroelectric dams (Wagner 1991; Bjornn and Peery 1992). Long et al. (1968) reported 30% combined turbine and predation mortality at Ice Harbor Dam. Reviews of turbine passage studies have concluded that turbine passage causes 10-30% direct mortality (NPPC 1986; Columbia Basin Tribes et al. 1993). Further, the stress of turbine passage may affect resistance to disease (Maule et al. 1988; 1989) and ability to avoid predators (Olla and Davis 1989).
River Ecology: The construction and operation of dams and cumulative impacts of mainstem habitat degradation have brought about many changes to physical and biological conditions to which salmon adapted, including changes in water quantity and quality (e.g., organic and inorganic sediment routing, turbidity and temperature) and the abundance of predator species (Junge and Oakley 1966; Li et al 1987; Petts 1980). Studies of predation in John Day Reservoir in the mid-1980’s showed that predators may have consumed 2.7 million juvenile salmon annually, with most of the consumption by big mouth minnows. Ruggerone (1986) estimated that gulls consumed 2% of the entire spring migration as they passed through the tailrace of Wanapum Dam. Changes in temperature conditions are addressed in the section on mainstem passage of adult salmon. As seen in other riverbasins worldwide, the loss of floodplain habitat by the operation of dams, dredging, and shoreline development may be a key factor limiting anadromous fish production and diversity (Dodge 1989).
Transportation: Mass collection and transportation of juvenile salmon by long haul barging and trucking has not offset the decline of Snake River salmon (Congleton et al. 1984; NMFS 1995) (Figure 5B.4). While numerous experiments to measure the effect of transportation have been conducted since 1968, because of poor study design, there is no definitive information to compare the cumulative and synergistic effects of transportation against historical passage conditions (e.g., spill, and increased flow), which supported sustained levels of stock productivity (Mundy et al. 1994). For example, the impact of juvenile transportation upon adults returning to the spawning grounds has never been fully investigated, although analysis by state, tribal, and federal salmon biologists suggested that the overall effect has been negative (TRG 1993).
It is likely that the transportation process (diverting juveniles into screened bypass systems, dewatering, sorting, handling, loading and unloading) is causing a selective mortality function to Columbia basin stocks as a whole. The transportation process has precluded the use of spill and additional flow to protect juvenile migrants. Yearling juveniles, which are larger and have undergone greater physiological development, tend to be diverted by screened systems in much greater numbers than subyearlings and fry. For example, fish guidance efficiency (FGE) at Lower Granite Dam for yearlings is estimated at about 60%, while FGEs for subyearlings are estimated at about 30%. Thus, about 70% of the subyearling run is subjected to turbine passage. As well, the transportation process selects against the subyearling life history when guided fish are subjected to extremely high water temperatures within screen bypass systems. For example, in 1994, approximately 100,000 subyearling chinook were lost in the McNary Dam screened system from thermal shock.
The transportation process also selects against the subyearling life history with respect to the timing of juvenile maturation. To achieve proper size at saltwater entry for ocean survival, juveniles must rear as they migrate downstream (Reimers 1973; Nicholas and Hankin 1988). Proper size at saltwater entry has been documented to be extremely important to survival to adult and age of maturation. For example, Reimers (1973) used scale analysis to find that returning adult fall chinook originated largely from juveniles that attained a size of 10-12 cm at saltwater entry. The transportation process interrupts rearing timing and places subyearlings in the lower Columbia in only a few days when under natural migration conditions, the juveniles would have utilized over a month to achieve the proper size for saltwater entry.
Historically, the subyearling life history dominated as the most abundant life history form basinwide. It is likely that this particular life history has been expressed to allow for stock resiliency in the face of changing environmental conditions (Lichatowich 1993; Schluchter and Lichatowich 1977). The selection against the subyearling life history by the use of the transportation process likely has and will continue to impact critical life stages such as freshwater and marine growth, age at maturation, fecundity, and spawner distribution. As these elements are profoundly impacted, peaks and troughs in stock abundance and the loss of diversity will become more extreme and may lead to extirpation.
Transportation has not resulted in improving smolt to adult survival rates to ranges necessary for restoration of basin salmon stocks. Unless these rates can be improved to historical ranges as documented by Raymond (1988), the issue of transportation is moot (Mundy et al. 1994).
In an adaptive management context, changes in hydrosystem configurations and operations intended to improve juvenile salmon productivity and improve survival must be sufficient in magnitude to produce a detectable response in salmon populations. A prescribed set of actions, taken in concert with one another because effects are interactive and cumulative, can result in a three to five-fold increase in survival for salmon stocks originating above eight or more dams. These actions include:
Water Flow: Adequate flows are necessary to restore mainstem habitat and promote life history diversity. Specifically, these will reduce smolt exposure time to adverse conditions, and facilitate their urge to migrate once triggered by physiological cues. Adequate flows also improve the effectiveness of spill, allow for more efficient turbine operations, reduce water temperatures, and improve bypass system performance. In the long-term, water flow actions should be based on an approach to recreate the natural hydrograph and reduce hourly and daily fluctuations due to power peaking (Figure 5B.5). With respect to daily and hourly flows, assume no more than a 10% variance in flow volumes at any specific point in the river in a 24 hour period. Provide flow minimums that are biweekly averages with weekend and holiday flows at least 80% of the previous five days’ flow levels.
Turbines: Operation of turbines to maximize hydraulic efficiency improves the survival of juvenile and adult salmon that pass through turbines during their migration. Structural and non-structural measures related to avoidance of turbine passage are discussed below.
Spill: Controlled spill improves passage survival of migrating salmon and steelhead and lamprey at mainstem hydroelectric projects. Spill allows juvenile migrations to pass hydroelectric projects without passing through turbines, reduces travel time through reservoirs, and reduces predation. In general, juveniles passed with controlled spill experience only 0-3% mortality (Holmes 1952; NPPC 1986; Raymond 1988; Ledgerwood et al. 1990). A program of controlled spills that maximize spill efficiency to allow 80-90% of juvenile migrants to pass each dam by nonturbine routes without creating unacceptable levels of dissolved gases should be implemented at all mainstem dams.
Transportation: Halting barging and trucking of salmon would allow testing of alternative passage measures, which have a greater potential for increasing passage survival. Alternative passage measures such as increased flows and spill have a proven historical basis (Raymond 1988) and are supported by studies of riverbasins worldwide (Dodge 1989) for increasing fish production.
Structural Considerations: Numerous studies have identified structural changes to improve juvenile salmon survival (USACE 1992-1995). Current structural features at the dams are inadequate for juvenile salmon passage. For instance, mortality to juveniles that travel through mechanical bypass systems ranges from 2 to 17% (NPPC 1986; WDFW 1991-1994). Major structural modifications at mainstem dams are needed. Permanent drawdown of certain Snake and Columbia river reservoirs should substantially benefit the survival of migrating juvenile salmon. Further development of surface oriented bypass systems in conjunction with spill are warranted (HTI 1995). Other needed modifications include changes to existing juvenile mechanical bypass systems and installation of fliplips at dams and other gas abatement structures where excess total dissolved gas from uncontrolled spill can cause gas bubble trauma. Measures such as load shifting must be instituted to deal with over-generation spill during times of high flows to allow operation of turbine units to control high dissolved gas levels resulting from otherwise uncontrolled spill.
Implement a program of short-term and long-term juvenile passage and mainstem habitat restoration measures at federal and nonfederal dams on the mainstem Snake and Columbia rivers. Utilize the historical Columbia basin and information from river basins worldwide as templates to establish protocols and to measure comparative progress. Establish a monitoring program jointly designed by tribal, state, and federal fish management agencies to estimate smolt-to-adult survival and individual stock production and fitness. Use noninjurious measurement techniques, such as sonar and scale-sampling procedures, to rigorously enumerate stock responses through precise dam and spawning ground counts. Emphasize measurements of life history characteristics such as time and size at important life history stages instead of complete reliance of studies of survival and abundance because the former are more statistically sensitive to change (Cramer and Lichatowich 1979; Lichatowich and Mobrand 1995). In-season management of juvenile passage measures should be based on cooperative efforts of tribal, federal, and state governments.
Water Flow: Implement instream flow measures (Table 5B.5) in the Snake and Columbia rivers. Flow measures for the mainstem Snake River are defined in terms of volumes of water to be provided for flow augmentation.* Flow measures for the Columbia River are defined in terms of minimum instream flows as measured at The Dalles Dam. Canadian reservoirs, comprising approximately one-half of the Columbia River’s storage, must be drafted as necessary to meet the flow targets at The Dalles. The targets change based upon reservoir conditions at the conclusion of each water year. These reservoir conditions generally reflect sequences of dry years, e.g., second critical year firm energy load carrying capability (FELCC) declarations generally reflect the second year of a two year drought. The sliding scale at The Dalles also includes a formula for increasing the flows based on storage capacities in above average water years, such that 40% of the runoff above average runoff volume at Hungry Horse and Libby dams would be provided in addition to the flow targets.
Provide the following Snake River flow augmentation volumes managed at the direction of the Columbia River treaty tribes and state and federal fishery agencies:
- Upper Snake–(1-3 maf)
- Brownlee–(450 kaf)
- Dworshak–(1.5 maf spring, 1.0 maf summer. To minimize impacts on Dworshak resident fish, wildlife, and cultural resources and allow use of Dworshak for summer temperature control), priority for meeting flow needs shall be achieved from Upper Snake River storage. Springtime operation of Dworshak should attempt to mimic natural runoff.
Turbines: Avoid operating turbines outside of 1% of peak efficiency. Completely avoid excursions during peak migration periods, particularly at projects that are experiencing large juvenile and adult migrations. Complete index testing of all river turbine units. Implement powerhouse optimization programs to improve turbine operating efficiencies at all dams. Examine the theory of increasing fish survival through increasing wicket gate openings.
Spill: Implement a program of controlled spill to achieve an 80% fish passage efficiency (fish passing by nonturbine routes), while managing spill so that dissolved gas concentrations do not exceed 125-130% daily average total gas pressure. Spill efficiency should be maximized through the use of hydroacoustic monitoring across the entire dam, installation of full flow surface bypass systems and implementation of evaluated spill patterns. Dissolved gas monitoring measures should be implemented as a part of this program to identify physical aspects of the gas plumes in the water column and to determine effects on fish in the river.
Predator Control: Continue evaluation of site specific intensive removal of predaceous bigmouth minnow. Implement evaluation of control programs for other predators including seagulls, bass, channel catfish, and walleye.
Transportation: Halt mass transport barging and trucking of juvenile anadromous salmonids from Snake and Columbia river dams.
Drawdown: Implement structural modifications necessary to lower the surface elevations of mainstem reservoirs in an adaptive management framework that includes biological, economic, and cultural study and mitigation measures agreed to by the Columbia River treaty tribes. Drawdown should be implemented on permanent (i.e., not seasonal) basis for ecosystem considerations. Implement the following measures as soon as possible:
- John Day Dam should be drawn down to minimum operating pool, elevation 257 msl, by the 1997 juvenile salmon migration;
- Lower Granite should be drawn down to elevation 710 msl by the 1997 juvenile salmon migration;
- Little Goose, Lower Monumental, and Ice Harbor dams should be operated at minimum operating pool from April 15 to October 31; and
- Engineering and related planning should be undertaken in cooperation with tribes for long-term drawdown options identified below.
Surface bypass: Expedite prototype development of surface flow bypass systems, particularly to address passage needs at Bonneville, John Day, The Dalles, Ice Harbor, Rocky Reach, and Priest Rapids and Wanapum dams.
Other structural measures: Install fliplips at John Day, Ice Harbor, Wanapum, and Rocky Reach dams as soon as possible. Immediately investigate gas abatement solutions at Bonneville Dam and install structural measures to remedy problems as a top priority. Evaluate the benefits of modifying the juvenile bypass outfall location at Bonneville Dam.
Cultural Resources Protection: Before proceeding with any reservoir actions that may expose Indian burial sites, village sites, or other Indian cultural resources of the Columbia River treaty tribes, the Corps of Engineers shall implement a plan for protection of these sites and resources, including mitigation of any impacts, which has been developed in cooperation with the affected tribe. The Corps of Engineers and the U.S. Department of Justice will enter immediate consultations with the tribes to address tribal cultural resource protection needs.
Monitoring and Evaluation: Monitor and evaluate salmon responses to restoration measures to the hydropower system by developing experimental and sampling designs for estimating total hydrosystem passage survival. Evaluate restoration actions by measuring changes to statistically sensitive life history parameters such as time and size of juvenile entry into saltwater and timing and distribution of adult spawners (Lichatowich and Cramer 1979; Lichatowich and Mobrand 1995).
Spill: Increase spill efficiency by installation of full flow surface bypass systems, gas abatement structures and modifications of spill patterns to achieve at least a 90% fish passage efficiency.
Drawdown: Several drawdown options with mitigation for biological, cultural and economic impacts agreed to by tribes and in conjunction with the other passage and habitat measures are set forth in this hypothesis. All drawdowns are year around in duration. The tribes’ preferred alternative for Snake River Dam drawdown would require structural modifications at Lower Granite, Little Goose, Lower Monumental, and Ice Harbor dams to allow for drawdown to natural river level. Drawdown to natural river level is generally intended to restore flows to the water surface elevations that existed in the Snake River prior to impoundment. John Day Dam drawdown to spillway crest should be considered a high priority. Alternative structural means of achieving these elevations are under consideration by the Corps of Engineers and HARZA. These alternative structural means include installation of low level water outlets at the dams, excavation of nonoverflow sections of the dams, and excavation of spillway sections of the dams. The following three options for permanent drawdown were analyzed for this hypothesis:
- John Day and Ice Harbor dams to natural river level;
- John Day to spillway crest, Ice Harbor and Lower Monumental Dam to natural river; and
- John Day and Snake River dams to natural river level;
In addition, implement drawdown at Wanapum and Rocky Reach dams. Besides reductions of passage and reservoir mortalities from these projects, substantial spawning areas for fall and summer chinook would be reestablished if these project were drawn down.
Flow Augmentation: In conjunction with drawdown of the Corps of Engineers’ Snake River and John Day dams, implement system operations to achieve mean historical flows during salmon migration periods (Table 5B.6). Historical flows in this context means those flows that would have existed prior to water resources development, including flows that would have occurred in the absence of irrigation depletions. To achieve these flows, dam operators should relax flood control rule curves to meet resident fish, wildlife, and salmon flow needs, and additional volumes should be obtained from Canadian storage reservoirs. Restore the natural hydrograph of the Clearwater River in the long-term.
Hells Canyon Complex: Take any actions necessary to restore salmon passage through the Hells Canyon complex.
Turbines: Retrofit existing turbines with more efficient turbine designs and automated operating procedures to decrease fish mortality. Reduce cavitation by eliminating fluctuations in forebay and tailrace elevations by restricting power peaking activities.
Modeling analysis conducted for purposes of this plan indicates that implementation of the long-term drawdown options in conjunction with the other measures set forth in the recommended actions/tests would increase survival of juvenile salmon originating from the Snake River above Lower Granite Dam by 3.0, 3.4 and 4.0 times recent survivals, respectively for options 1, 2, and 3. Modeling analyses by the states and tribes involved in the IDFG v. NMFS processes indicate a high probability of survival and recovery of spring, summer, and fall chinook populations under a scenario that calls for natural river drawdown of the Snake River dams and spillway crest drawdown of the John Day Dam. This alternative is similar to option 3 and should result in survival increases that fall between options 2 and 3. If drawdown and flow actions are not implemented and conditions of the past ten years are assumed into the future, modeling analyses show that Snake River spring, summer, and fall chinook are likely to be extirpated. Data collected will be used to improve and refine our understanding of smolt survival through the hydrosystem and to improve management tools.
Cooperative approaches among federal, state, and tribal governments are needed to ensure the success of these measures. Among other things, the Corps of Engineers will play a central role in development and application of technologies such as reservoir drawdown and the implementation of research programs intended to assess the effectiveness of fish passage measures. These technology development and research efforts should proceed with the full involvement of state and tribal governments and non-federal dam operators. Furthermore, implementation of fish passage measures, particularly including flows and spills, benefits from cooperation among federal, state, and tribal governments. In the case of measures at nonfederal dams on the Columbia River, the cooperative approaches now in place should serve as a model for the measures at the federal dams. These institutional recommendations are discussed more fully in the institutional and legal section of the plan.
*The instream flow measures, including augmentation volumes and flow targets, presented in this plan (both short-term and long-term) are not intended in anyway to supersede or set a precedent with regard to claims filed by the United States or the Nez Perce Tribe in the Snake River adjudication or set a precedent for claims for reserved water rights intended to protect the tribes’ treaty and aboriginal rights.