Passage and Survival of Juvenile Chinook Salmon Migrating from the Snake River Basin

Proceedings of a Technical Workshop University of Idaho

February 26-28, 1992

Presented by

Idaho Chapter of the American Fisheries Society Idaho Water Resources Research Institute, University of Idaho

Idaho Cooperative Fish and Wildlife Research Unit College of Forestry, Wildlife and Range Resources, University of Idaho

W estem Division of the American Fisheries Society

These proceedings have not been peer-reviewed.

145

Survival of Juvenile Salmon PassingPas.sing through Bonneville Dam and Tailrace

E. M. DAWLEY, R. D. I...EDGERWOOD, L. G. GU.BREATH, B. P. SANDFORD,P. J. BENTLEY, M. H. ScHIEWE, ANDS, J. GRABOWSKI

Na1ional Marine Fisheries Service, Northwest Fisheries Science Cenler 2725 Monllake Boulevard East, Seattle, Washington 98112

Introduction

The efficacy of bypass systems as a safe means of passing juvenile salmon (Oncorhynchus spp.) around dams has generally been established by site-specific testing as each system goes into operation. Virtually all of this testing focused on the portion of the system beginning at interception of out-migrant juvenile salmonids by submersible travelling screens, and ending just prior to release into a conduit or channel through which fish were carried to the tailrace below the dam. There have been few rigorous assessments of fish survival through an entire bypass system, from forebay to tailrace or beyond, at any of the dams.

The principal constraint in conducting such tests is the difficulty of obtaining a statistically unbiased sample of fish exiting a bypass system prior to reaching the next dam downstream. Assessment downstream is often complicated by the uncertainties of collection efficiency. This is not the case at Bonneville Dam, the lowermost hydroelectric project on the Columiba River, where approximately 157 km of free-flowing river separate the dam from the estuary. An established sampling station is located at the head of the estuary at Jones Beach (River Kilometer 75; Figure 1). More than 20 years of sampling, using beach and purse seines to collect out-migrating juvenile salmon, has demonstrated that an unbiased estimate can be made at this site (Dawley et al. 1986)

There are several compelling reasons for focusing research on fish passage at Bonneville Dam. As the last dam on the Columbia River, Bonneville Dam is in the critical position of passing more juvenile salmon than any other dam in the system. Moreover, no thorough assessment of passage survival has been conducted at the dam since completion of the spillway flow deflectors in 1975, the Second Powerhouse in 1983, and the two downstream migrant bypass systems in 1981 and 1984, Information specific to each of these separate passage routes is needed for management of fish passage relative to power production.

Passage Route Survival Comparisons

In 1987, the National Marine Fisheries Service (NMFS), in cooperation with the U.S. Army Corps of Engineers (COE), began a multiyear study to evaluate survival of subyearling fall chinook salmon O. tshawytscha passing Bonneville Dam. During June, July, and August, 1987 through 1990, groups of differentially marked chinook salmon were simultaneously released to pass Bonneville Dam via the spillway, the Second Powerhouse turbines, or the Second Powerhouse bypass system (Figure 2). Additional releases were made in the tailrace at the downstream edge of the turbine boil, about 2 km downstream from the dam. To date, about 8 million fish have been released. Estimates of short-term relative survival were based on recoveries of juveniles by beach and purse seines at Jones Beach. Estimates of long-term relative survival Were based on recoveries of tagged adult fish from the fisheries and from hatchery escapement.

The most striking finding of this study was that differences in estuarine recoveries of juvenile salmon from turbine and bypass release groups suggested little survival benefit associated with the bypass system. In 1987 and 1988, recoveries of bypass-released groups were significantly less than recoveries of turbine­released groups; mean differences were 13.3 and 16.6 %, respectively. In 1989 and 1990, recoveries of bypass-released groups were also less than recoveries of turbine-released groups, though not significantly; mean differences were 3.1 and 2.6%,respectively. The difference between data sets may be associated with r river flow, resulting in higher tailwater elevation during tests conducted in the last 2 years (Figure 3).

1:J

N (j ' I

-· O· 25km � Washington -.

(j

0 C')

(t)

� Jones Beach

::i Oregon RKm 75

• I

,Jr! \-��

Bonneville Dam J'-1 I .... ),>, I

RKm 232 ,::1 '"""' � I

. �� '� . I l .. ;..

� wasnington �-r' I

I

r<-i_".J.-!'

Figure 1.

. \ ___ ..

The lower Columbia River showing locations of Bonneville Dam and the estuarine sampling site at Jones Beach, Orgeon.

146

Hamilton Island boat launch

// •

Downstream release

Figure 2.

�Second Powerhouse �

Upper turbine, lower turbine, and bypass system releases

Spillway release

Forebay

I • Release locations !

Schematic of Bonneville Dam and vicinity showing release locations for subyearling chinook salmon during 1987-90 studies.

147

o/c;,

s u

r V

i

V

Figure 3.

-- % survival � T.W. elevation (m)

100 .------------------. 7

95

90

85

6 T a

5 w

a t e

4 r

e I

3 e V

a t

2 i ········································································································ 0

n

1 m

80'---------L--___ __._ _____ 0 1987 1988 1989 1990

Increased survival of bypass-released groups relative to increased tailwater surface elevation; survival % = (Bypass recovery %) / (Lower turbine recovery % x 100). Only the last half of the annual releases were used to provide uniform survival data for comparing low tailwater test conditions.

148

Higher tailwater elevation has diminished turbulence and reduced water velocity from 8.1 to 7 .6 m/sec within the 1-m diameter bypass conduit, In addition, it has diminished shear forces at the conduit terminus.

Comparisons of recovery differences between bypass-released and other release groups were also made, but include fewer years of comparison (Table 1). On the basis of 3 years of releases, recoveries of bypass-released groups averaged 8.3 % less than recoveries of tailrace-released groups. From 2 years of releases, recoveries of bypass-released groups averaged 17 .4 % less than recoveries of downstream-released groups. On the basis of data from a single year {1989), bypass-released groups averaged 16.6% less than spillway-released groups. This latter comparison is noteworthy since spillway passage has long been believed to be the safest route of dam passage and was considered similar to bypass passage.

Although several years remain before data on adult returns are complete, preliminary results suggest that returns of bypass-released fish are not significantly different from those of turbine-released fish. Again, this suggests a lack of benefit from bypass passage (Ledgerwood et al. 1990, 1991).

Evaluation of the Bypass System Results of passage survival tests prompted us to focus research efforts on detrimental impacts to out-migrating juvenile salmon using the bypass system. Decreased survival may b;e a consequence of physical damage, occurring during passage through the system; increased predation after egress from the bypass discharge conduit; or a combination of both.

The design and location of the bypass conduit terminus were engineered to provide out-migrants the best possible protection against predation by birds and fish. The supporting structure for the 1-m diameter conduit is a teardrop-shaped column projecting 9 m upward from the river bottom, the top of which is 7-14 m below the water surface. The terminus is located in relatively high-velocity water about 76 m downstream from the dam, 85 m from the north shore, and 30 m downstream from the turbine discharge boil. The river bottom is smooth and the distance from any geologic relief was thought to eliminate predator sanctuary near out-migrating juvenile salmon.

Initial Investigations

Initial investigations of the physical features of the bypass system by NMFS and COE provided little evidence of problems. A video inspection of the discharge conduit revealed no structural problems sufficient to cause injuries to fish. At operating conditions identical to those of survival tests, water velocities adjacent to the discharge monolith varied from 1 to 1.6 m/sec, similar to model-predicted velocities. Northern squawfish Ptychocheilus oregonensis are thought to be the primary piscivore on juvenile salmon in the Columbia River (Poe et al. 1991; Vigg et al. 1991). Literature regarding habitat suitability for northern squawfish suggested that velocities of that magnitude would be exclusionary (Faler et al. 1988). Purse seining at the bypass outlet produced little evidence of injury or mortality, but insufficient fish were recovered to allow rigorous assessment.

Trap-Net Method

In 1990, researchers began using a trap-net recovery system to assess the physical condition of out-migrants following passage through the bypass system (Figure 4). The trap net was attached directly to a steel carriage permanently affixed to the outlet monolith. The objective was to identify which segment of the system was detrimental to juvenile salmonids and to assess any differences at various tailwater elevations. Marked hatchery and run-of-the-river fish were released at various locations in the bypass system and trap net. Fish were then recovered and evaluated for stress, scale loss, injury, mortality, and 48-h delayed mortality. To assess whether the trap net captured live and moribund fish in the same percentages, recently killed fish were released through the bypass system in conjunction with live fish.

149

TABLE 1.-Differences in relative survival between fish passing through the bypass system and other passage routes at Bonneville Dam based upon juvenile recovery data from estuarine sampling.

Percent difference of bypass recoveries from indicated treatmenia Release site/ passage route 1987 1988 1989

Turbine: Released at the ceiling and mid-depth of the turbine intake. Passage through the turbine and through the PH-2" tailrace.

-10.8• -13.6• -3.3

Tailrace: Released at the downstream side of turbine discharge boil. Passage through the PH-2 tailrace.

-14.1• -7.3

Spillway: Released 0.5 m above spillway crest.

1990

-2.5C

-3.6

Passage over the spillway, through stilling basin and spillway tailrace.

-16.6•

Downstream: Released downstream from dam and tailraces, at a swift-water site.

-23.1* -11.6•

• Calculated using annual means for recovery percent of treatment groups, where:

Average

-7.6•

-8.3•

-16.6*

-17.4•

BY = bypass, and TR == other treatment groups/passage routes [(BY% - TR%) + TR%] x 100.i. Abbreviation for Second Powemouse.c Only the mid-depth release site was used, to provide increased numbers of replicates.• Significant at P = 0.05.

150

� "'-:"

.,_

....

-·

-

.......

· Figure 4.

Bypass Collection Channel

ur1n --=11 1··

R ._ 11· _,

0 - -1 - 11-

1-�:.1-

Ofscf, ..

rnhrn9n � � hoso9 (19 cm)

uroo Cond I .

Ufffrnc11a.)'Li__./ l

�J

Fish !odder dlschoroo

2 TRAP-NET

release site

Flow

OUTLET MONOLITH 1--4 (olcv. -4.0 m)

Schematic of the downstream migrant bypass system at Bonneville Dam Second Powerhouse; fish release sites are numbered.

Trapping Results

The trap net recovered 80..100% of live test fish and 93-100% of killed fish released into the bypass channel. The high recovery rate of killed fish allowed us to assume that injured, moribund, and dead test fish exiting the bypass were proportionally represented in recovery data.

Tests conducted during the juvenile migration period at moderate tailwater elevations (4.9-6.1 m) showed some impacts from bypass passage, primarily increased stress and scale loss, with slight increases of injury and mortality. However, some groups showed substantial impacts from bypass passage. Run-of-the-river yearling and subyearling cbinook salmon and coho salmon 0. kisutch incurred high percentages of scale loss (9-29 % ; Figure 5) and individual fish suffered severe scale loss. In tests with subyearling chinook salmon, hatchery fish incurred 0.3 % injury, 5.6 % direct mortality, and 2.1 % delayed mortality; and run-of-the-river fish incurred 3.8% injury, 2.4% direct mortality, and 7.6% delayed mortality.

Assessment of blood plasma cortisol, glucose, and lactate-used to evaluate stress-indicated significantly greater stress to bypass-released fish than to controls. Concentrations of plasma cortisol were high for an 18-h period following passage (Figure 6). Test fish stressed in the laboratory by dip netting to establish comparison points of known stress showed cortisol levels of similar magnitude to those produced from bypass passage.

Additional tests conducted at low tailwater elevations (2. 7-3.2 m) may indicate high variability in passage conditions, causing intermittent high mortality. The high velocity and turbulence within the system during tests was similar to conditions that migrants would encounter during years of low river flow. Mortality ranged from 6 to 51 % for subyearling chinook salmon and from 0 to 32 % for subyearling coho salmon (Figure S). Impacts to fish released at the midpoint of the bypass system were less, 2-40 % mortality for subyearling chinook salmon and 0..19% mortality for subyearling coho salmon.

Two potential sources of hazard within the bypass discharge conduit were identified by NMFS biologists and engineers: entrained air may be causing severe pressure fluctuations throughout the conduit, and a short­radius elbow at the upstream end of the conduit may be producing negative pressures. Biological evaluation is in progress.

Aspects of Bypass Systems That May Cause Decreased Survival

Coincidental studies of northern squawfish in the tailrace of Bonneville Dam indicated greater predation on fish leaving the bypass system than at other locations at the dam. Ward et al. (1992) stated that trolling with lures at the bypass outlet produced substantially higher catches of northern squawfish than at any other location in the forebay or tailrace of the dam. In 1990, passage survival tests indicated that northern squawfish consumed .greater percentages of bypass-released fish than tailrace- or turbine-released fish (1bomas Poe, unpublished report, U.S. Fish and Wildlife Service, Columbia River Field Station).

Reduced survival of test fish released through the Bonneville Dam Second Powerhouse bypass system in 1987-90 probably resulted from both physical impacts of passage through the bypass discharge conduit and predation during migration through the tailrace. Physical problems within the bypass conduit will be remedied insofar as possible, but since the conduit is 287 m long, mostly 1 meter in diameter, and partially submerged, it may be difficult to identify and correct all problem areas. However, the inherent impact from predation following egress from the system cannot easily be remedied.

Predation may be an insurmountable problem as a result of (1) increased stress from passage causing diminished avoidance reactions [laboratory studies showed that severe stress or severe turbulence caused loss of equilibrium and abnormal avoidance behavior (Groves 1972; Sigismondi and Weber 1988)];

152

IMPACTS FROM BYPASS PASSAGE

- Descaling [2J Mortality

+ Tail Water Elevation

60 6

50 5

p

e

r C T

e 40

n t

0 e

f 30 3

T t

e I

s 0 t

20 2 n

F m I

s

h 10 1

0

-'9--9 'Y IS' 0

'9'9 ;,IS'};. 'J,

C'o C'�. 0 'J,o �o //) �/.; 0

,1 0

Figure 5. Average descaling and mortality percentages of test fish in relation to tailwater (IW) elevation during tests (RR = run-of-the-river, H = hatchery, Yr = yearling, Sy =subyearling, and Chin = chinook salmon).

153

PLASMA CORTISOL

400 ---·-·········································································

River Run Yr Chinook //R,"

R /It V

Figure 6.

n

g I

m

I

300 - - ............................... . . .... / ... .. -:-:-.>.ef-. - .. :., .. ................ .

/ / Lab Chronic

Lab Acute

0

-0.5 0.5 2 4 6 18 42

Hours (relative to test)

400 ---·····································--·····--················· ·············· .................. .

River Run Subyr Chinook / / ,5751---V"

n

g I

m

I

300 - · · .. . ... · ... · .. · · · · · -/· · - · ..... . /v

/

/ /

200 -- ... ... ... ····/"'"

100 --

0

�

/

/'.

-0.5 0.5 2 4 6 18 42

Hours (relative to test)

Cortisol levels of yearling and subyearling chinook salmon before and after bypass passage compared to counterparts released into the trap-net (Net) and to laboratory test fish stressed by dipnetting (Acute) and by continuous crowding (Chronic).

154

(2) point-source release from the bypass allowing predators to congregate; 3) migration through a low-velocity tailrace basin providing a large area of suitable habitat for northern squawfish; and (4) a bypass outlet location on the north side of a tailrace that angles to the south about 90°, tending to direct out-migrants shoreward toward rip-rap areas-prime habitat for northern squawfish.

In the passage survival tests of 1987-90, estimated survival for test fish released 2.5 km downstream from the dam was significantly higher than for fish released into the bypass system. The physical conditions at the downstream release location that most probably allowed higher survival were (1) high water velocity-1.5 to 2.1 m/sec; (2) long distance from shore-about 100 m; (3) rapid downstream dispersal of fish, resulting in decreased juvenile salmon density in the migration route and increased time for orientation prior to encountering predators; (4) release where current ditection was parallel to the shoreline; (5) laclc of predator attraction from a continuous egress of juvenile salmon at a single location or along a localiz.ed migration route; and (6) nighttime releases that minimized avian predation.

Summary and Conclusions

1. Trends observed in the juvenile recovery data suggest that bypass passage has not substantiallyimproved survival as compared to turbine passage for summer-migrating juvenile cbinook salmon atBonneville Dam; however, the final conclusions regarding differences in,passage survival must awaitanalyses of all adult returns.

2. Bypass passage appears to cause significant stress, loss of scales, and some direct mortality. Duringsummertime low river flows, the resulting low tailwater elevations appear to aggravate mortality duringpassage.

3. Survival of fish leaving the bypass system appears to be diminished by northern squawfisb predation.

4. Conditions that appear to increase survival of downstream- released fish over bypass-released fishinclude high water velocity, long distance to predator habitat, current direction parallel to the shoreline,low level of stress for migrants at river entry, lack of predator attraction from continuous availability ofjuvenile salmonids, and nighttime release of fish to limit avian predation.

5. Conditions thought to decrease survival of out-migrating juvenile salmonids at the Bonneville DamSecond Powerhouse bypass system may be important at other dams, and should be investigated.

References

Dawley, E. M., R. D. Ledgerwood, T. H. Blahm, C. W. Sims, J. T. Durkin, R. A. Kim, A. E. Rankis, G. E. Monan, and F. J. Ossiander. 1986. Migrational characteristics, biological observations, and relative survival of juvenile salmonids entering the Columbia River estuary, 1966-1983. Report to Bonneville Power Administration (contract DE-A179-84BP396S2). (Available from Northwest Fisheries Science Center, 2725 Montlake Blvd. E., Seattle, Washington 98112-2097.)

Faler, M. P., L. M. Miller, and K. I. Welke. 1988. Effects of variation in flow on distributions of northern squawfisb in the Columbia River below McNary Dam. North American Journal of Fisheries Management 8:30-35.

Groves, A. B. 1972. Effects of hydraulic shearing action on juvenile salmon. Unpublished manuscript. (Available from Northwest Fisheries Scieace Center, 2725 Montlake Blvd. E., Seattle, Washington 98112- 2097.)

155

!i

Ledgerwood, R. D., B. M. Dawley, L. 0. Gilbreath, P. J. Bentley, B. P. Sandford, and M. H. Schiewe. 1990. Relative survival of subyearling chinook aalmon which have passed Bonneville Dam via the spillway or the Second Powemouse turbines or bypass system in 1989, with comparisons to 1987 and 1988. Report to the U.S. Army Corps of Engineers (contnct B8S890024/B86890097). (Available from Northwest Fisheries Science Center, 27'15 Montlake Blvd. E., Seattle, Washington 98112-2097.)

Ledgerwood, R. D., E. M. Dawley, L G. Gilbreath, P. J. Bentley, B. P. Sandford, and M. H. Schiowe. 1991. Relative survival of subyearling chinook aalmon which have passed through the turbines or bypass system of Bonneville Dam Second Powerhouse, 1990. Report to the U.S. Army Corps of Engineers (contract B86900104). (Available from Northwest Fishenes Science Center, 27'15 Montlake Blvd. B., Seattle, Washington 98112-2097.)

Poe, T. P., H. C. Hansel, L. Vigg, D. B. Palmer, and LA. Prendergast. 1991. Feeding of predaceous fishes on out-mi,rating juvenile salmonids in John Day Reservoir, Columbia River. Transactions of the American Fisheries Society 120:405-420.

Sigismondi, L A., and L J. Weber. 1988. Changes in avoidance response time of juvenile chinook salmon exposed to multiple acute handling stress. Transactions of the American Fisheries Society 117: 196- 201.

Vigg, S., T. P. Poe, L A. Prendergast, and H. C. Hansel. 1991. Rates of consumption of juvenile salmonids and alternative prey fish by northern squawfish, walleyes, smallmouth bass, and channel catfish in John Day Reservoir, Columbia River. Transactions of the American Fisheries Society 120:421-428.

Ward, D. L, M. P. Zimmerman, R. M. Parker, and S. S. Smith. 1992. Development of a system-wide predator control program: indexing, fisheries evaluation, and harvesting technology developmenL In C. F. Willis and A. A. Nigro, editors. Development of a system-wide predator control program: stepwise implementation of a predation index, predator control fisheries, and evaluation plan in the Columbia River Basin, 1991. Bonneville Power Administration, Portland, Oregon.

156