Review of the U.S. Army Corps of Engineers' Capital Construction Program:
Part II.A Development and Testing of Surface Bypass
September 29, 1998 | document ISAB 98-7
Review of the U.S. Army Corps of Engineers' Capital
Construction Program: Part II.A Development and Testing of
Richard N. Williams, Chair
Peter A. Bisson
Charles C. Coutant
Phillip R. Mundy
Richard R. Whitney, Cochair
ISAB RECOMMENDATIONS ON SURFACE BYPASS
Migratory Behavior of Juvenile Salmon
Bypass for Juvenile Salmonids at the Dams
Turbine Intake Screens
Spill for Fish Passage
Surface Flow Bypass Development
Tests of Prototype Surface Bypass Collectors
Premises for Surface Bypass Development
The Northwest Power Planning Council (the Council) has been directed by
the Congress, in the Conference Report accompanying the Energy and Water
Development Appropriations Act for the Fiscal Year 1998, to review
"the major fish mitigation capital construction activities proposed
for implementation at the Federal dams in the Columbia River Basin."
The Council was directed to conduct this review by June 30, 1998, with the
assistance of the Independent Scientific Advisory Board (ISAB).
The Council identified the Corps? development and testing of the
surface bypass prototype at Lower Granite Dam and at other dams, hereafter
termed "surface bypass," or "SBC," as one of five
specific projects requiring attention of the ISAB. Other specific projects
included installation of extended-length screens at John Day Dam and at
other dams, the dissolved gas abatement program, Bonneville Dam juvenile
fish passage improvements including the relocation of a bypass outfall,
and adult fish passage improvements.
This report is the second part of a series developed by the ISAB to
assist the Council in responding to the congressional mandate. Our
previous report, ISAB 98-4, submitted in May, 1998, covered questions from
the Council about the ecosystem context for mainstem fish bypass measures,
about proposed installation of extended-length turbine intake screens at
John Day Dam, and the proposed relocation of the juvenile fish bypass
outfalls at Bonneville Dam. The ISAB requested an extension of time, to
September 1998, to develop this response to questions regarding
development of surface bypass for juvenile salmonids and abatement of
supersaturated gas caused by hydroelectric project operations. Questions
on the Corps' gas abatement project are addressed in a companion report
(Part II. B. ISAB 98-8). We were aided in our review by a briefing by the
Corps and concerned agencies on July 15, 1998 and numerous reports and
reprints of published literature.
Due to the amount of time available for preparing this report, the ISAB
focused on the immediate question of whether the surface flow bypass
technology shows sufficient promise to warrant continuing development
under the existing configuration and operation of the hydroelectric
system. The more detailed suite of questions posed by the Council in its
charge to the ISAB in January of this year will be addressed in the final
report of the ISAB on the Corps? capital construction program in January
The immediate question before the ISAB is, "Do the preliminary
tests in prototype indicate that the surface flow bypass technology shows
sufficient promise to warrant continuing development?" Under the
presumption that the system of hydroelectric dams retains its present
configuration and operations, the answer is "Yes". It is
important to note that this is not a blanket endorsement of surface bypass
technology, since the efficacy of surface bypass technology will need to
be established for each individual dam. The epitome of successful surface
bypass is known as the Wells Dam concept. However, as a consequence of
structural differences between Wells Dam and the other dams, the Wells Dam
concept will need to be adapted to the configuration of each dam through
development and testing of a prototype.
ISAB RECOMMENDATIONS ON SURFACE
The following summary recommendations on surface bypass will be more
fully developed in the final report on the Corps? capital construction
projects to be delivered in January 1999.
- The Corps should continue development and testing of surface
bypass prototypes in consideration of the following recommendations.
- Careful consideration needs to be given to the locations within
the hydroelectric system where surface bypass would do the most good.
- Priority needs to be given to understanding the effect of surface
bypass on federally listed species, however the effects of surface
bypass on all species and life history types of native fishes needs to
- Development needs to take into account the need to protect the
widest possible biological diversity.
- An aggressive, nontraditional approach to prototype development,
which involves fast-track design, construction and testing, should be
- A time period of approximately ten years should be provided for
the developmental phase of surface flow bypass systems to ensure
proper and full evaluation.
- Testing should focus on developing data to evaluate the basic set
of premises or hypotheses on which surface flow bypass is based.
- A set of alternate hypotheses should be articulated for future
testing in the event that any of the basic hypotheses are rejected.
- Development of surface bypass and development of gas abatement
measures are closely related and they therefore need to be consistent
with each other.
In Return to the River (ISG 96-6) the Council's Independent
Scientific Group undertook a detailed, thorough review of bypass measures
being undertaken for juvenile salmon and steelhead at mainstem Columbia
River and Snake River dams. The chapter dealing with that subject was
subsequently revised and issued separately by the Council (Whitney et al.
Evaluation of the Corps of Engineers projects to develop surface bypass
facilities (collectors) for juvenile salmonids requires an understanding
of the basic problem whose solution is being sought, as well as an
appreciation of work that has led up to the decision to attempt the
development. Development of surface bypass was first mentioned as a
recovery tool for threatened and endangered salmon in 1993 by the Snake
River Salmon Recovery Team in their first report to NMFS (SRSRT 1993; NOAA,
1995). In a subsequent report to the Northwest Power Planning Council,
Harza presented specific surface bypass design concepts, including cost
estimates, with a recommendation to proceed with testing at Lower Granite
Dam (Harza 1994). The Independent Scientific Group made a recommendation
for surface bypass in 1996, concluding, "Surface collectors are the
most promising devices for attaining the fish passage goals established by
the council in the FWP or NMFS/NOAA in the Snake River Salmon Recovery
Plan." (ISG 96-6, p. 308). In addition the Council's Fish and
Wildlife Program calls upon the Corps and the mid-Columbia P.U.D.s to
explore new approaches for fish bypass technology, including surface
bypass systems, and surface spill. See Whitney et al. (1997) for further
Migratory Behavior of Juvenile Salmon
In addition to the reports cited above (Whitney et al. 1997 and ISG
96-6) the review by Johnson and Dauble (1995) is particularly valuable as
a basis for the following discussion on migration. As they move downstream
in the mainstem Snake and Columbia rivers, juvenile salmon are located
primarily in the upper portion of the water column, mostly within about 30
to 35 feet of the surface. When they encounter a dam, they can pass by
several routes. The primary routes for passage are the turbine intakes or
the spillway. Because the turbine intakes extend from near the river
bottom upward usually to within about 35 feet of the surface, the route
through the turbines requires that the bulk of the fish move downward to
enter the turbine intakes. Evidence has accumulated showing that passage
through turbines occurs mostly at night. On the other hand, when surface
passage routes exist, the fish use them rather uniformly throughout the
Differences exist in migratory behavior among the salmon species and
among stocks with different life history patterns. There have been
apparent shifts in the annual timings of the juvenile emigration, as a
whole, due to changes in relative abundance of the life history types,
which can have different migratory timings. At the present time, the bulk
of the emigration occurs in the spring months (April, May and June). While
some subyearling chinook are present at that time, they also show an
increase in abundance later in the summer (July and August).
A final but crucial aspect of migratory behavior in juvenile salmon is
the tendency to follow the bulk flow as it approaches the powerhouse.
Juvenile salmon tend to be in the upper 30 to 35 feetof the water mass and
the direction and velocity of its movement direct the movements of the
juveniles. The tendency to follow bulk is a factor in the successful
surface bypass at Wells Dam (Mike Erho and Gary Johnson, personal
communication). See the section below on Wells Dam for further
Bypass for Juvenile Salmonids at the
Turbine Intake Screens
When the hydroelectric system on the mainstem Columbia and Snake rivers
was under development, provision was made for upstream passage of adult
salmon and steelhead, but there was not a universal understanding that
there was a need to provide passage facilities for juveniles. It was not
until development was well underway that studies demonstrated that
juveniles passing through the turbines experienced substantial levels of
mortality, in the range of 11% to 15% in the first studies. The Corps and
other developers were faced with the challenge of developing a technology
that would reduce this level of loss, and of retrofitting dams that were
already built. Investigators at the National Marine Fisheries Service
developed a concept of a traveling screen at the entrance to the turbine
intakes that would divert the fish upward into gatewells where they could
be collected and routed around the dam. Prototype designs were tested
beginning in 1969. When Lower Granite Dam was built in 1975, it included a
full array of intake screens. The technology has been improved over the
years, and continues to improve as tests of various prototype designs have
been conducted. The measure of success is the percentage of fish
approaching the turbine intake that are successfully diverted by the
screens. This measure is the fish guidance efficiency (FGE) of the
screens. All four of the Snake River dams are now equipped with turbine
intake screens, as are all but The Dalles Dam in the lower Columbia River.
Three of the eight Snake-Lower Columbia River dams have extended length
turbine intake screens (Lower Granite, Little Goose and McNary). In the
mid-Columbia Reach, only Wells Dam has a fully functioning bypass system
for juvenile salmon. It is a surface collection system, not an intake
screen system. We will describe it in the text below.
Effectiveness of turbine intake screens seems to have approached an
upper limit where large increases in the fraction of juveniles diverted
are not likely to occur. No known type of bypass system is completely
effective at keeping juvenile salmon away from the turbines at operating
hydroelectric dams. Even when taken to the limits of available technology,
intake screens are unlikely to prove 100% effective in diverting juvenile
salmon (Office of Technology Assessment, 1995, p. 127). Indeed, intake
screens have not so far been able to achieve the goal of 80% diversion for
all federally listed species (Whitney et al. 1997).
Spill for Fish Passage
The deliberate use of spill as a passage route for juvenile fish has
evolved since 1980, when a FERC agreement called for a specified
percentage of river flow to be spilled during the spring emigration at the
P.U.D. projects in the mid-Columbia Reach. When the Council adopted its
first Fish and Wildlife Program it included provisions for spill during
the spring. Currently, the 1994 Fish and Wildlife Program of the Council
calls for 80% fish passage by safe routes, which includes fish diverted
from the turbine intakes into a bypass system, as well as fish in spill.
The Biological Opinion of NMFS/NOAA contains the same standard. Because
the turbine intake screens alone will not accomplish the 80% standard for
all species, spill must be added in sufficient amounts to supplement what
the screens can achieve.
Implementation of this 80% fish passage efficiency (FPE) standard is
complicated by the difference in effectiveness of the screens for
particular species and life history types of the fish, and by differences
among individual dams. Generally, the screens are most effective for
yearling chinook, coho and steelhead, less so for sockeye, and least
effective for subyearling chinook. To achieve the 80% fish passage
standard, spill must be added to supplement the FGE of the screens. To
determine a level of spill achieving the 80% fish passage standard
involves settling upon a composite FGE percentage derived from the
different percentages of the particular fish expected to be diverted by
the screens. A separate composite percentage is set for each dam in the
Snake and Lower Columbia River. In the spring, since the large majority of
the fish show high values of FGE for the screens, the composite FGE number
is heavily weighted in their favor, to the disadvantage of sockeye and
subyearling chinook. This problem becomes somewhat academic in the final
analysis because the spill amounts required could not be achieved because
of water quality standards that limit gas supersaturation produced by
spilling the requisite amounts of water. In 1998 in response to this
dilemma, NMFS shifted emphasis from requiring that the 80% fish passage
standard be met, to setting spill levels at each project that are expected
to meet the maximum 120% gas saturation standard (Gary Fredericks, NMFS
Portland, personal communication). Note that in the case of the
mid-Columbia P.U.D. dams, spill amounts are set through a FERC process.
Surface Flow Bypass Development
The natural behavior of juvenile salmon, previously described, places
them predominantly in the upper portion of the water column. As
alternatives to turbine intake screens are considered, developers hope to
make use of this characteristic behavior. It has been observed that where
spill can be drawn from the surface, it is more effective in attracting
fish than spill drawn from deeper water. Modification of spill gates is
being investigated as a possible means of improving fish passage
efficiency while maintaining spill volumes at levels that will meet water
quality standards. Preliminary tests with sluiceway gates that were
already designed to draw from surface waters have shown promise.
The primary source of encouragement for surface flow bypass development
is the success at Wells Dam in the mid-Columbia Reach. Wells Dam includes
the first successful surface flow bypass system for juvenile salmon.
Testing of a prototype began in 1983. Full installation across the
powerhouse was complete in 1989. Studies over the next three years showed
that the bypass was successful in passing 89% of the juvenile fish that
passed the dam, both in spring and summer (Skalski, 1993). To date, it is
the only system in the Columbia Basin that achieves the 80% fish passage
standard without the addition of spill.
One of the reasons for the success of surface collection at Wells Dam
is the particular design of the dam itself, known as a hydrocombine, in
which the spillway is located directly above the turbine intakes.
Hydraulic model studies indicated the feasibility of the concept. In the
prototype and final design, solid baffles were placed in front of the
spillway entrances to a point 30 to 40 feet below the surface. In five of
the eleven enclosed spillbays a vertical slot was left open for passage of
water and fish in the center slot of those spill bays. Opening the top
leaf of a spillway gate provided fish attraction flow that kept the fish
in the upper strata of the water column, rather than sounding to pass
through the turbine intakes (Johnson, 1995).
Another reason for the success of surface bypass at Wells Dam is that
juvenile salmon generally follow the bulk flow into the forebay. Juvenile
salmon are usually found in the upper 30-35 ft of the flow approaching the
1000-ft width of the hydrocombine. The turbine units draw from deep in the
water column, and the lines of flow into the turbines are relatively flat.
Hence the combination of fish distribution and hydrocombine hydraulics
make it likely that the juvenile salmon will encounter one of the surface
bypass entrances at Wells Dam.
The design and configuration of the entrances also play important roles
in the success of Wells Dam surface bypass. Placement of the five
entrances across the powerhouse is designed to ensure encounters by
juvenile salmon. In addition to location, three factors are key to the
success of the entrances; large size, appropriate acceleration coefficient
(Arnold 1974), and high volume of flow relative to total discharge (Gary
Johnson and Mike Erho, personal communications).
Tests of Prototype Surface Bypass
The surface bypass technology developed at Wells Dam is not directly
transferable to other dams on the Snake River or mainstem Columbia River.
The design of Wells Dam, the hydrocombine, is fundamentally different from
the design of all other Snake and Columbia mainstem dams. Hydrocombines
have the spillway located directly above the powerhouse, whereas the other
mainstem dams have spillways located separately from the powerhouse.
It would be premature to review the details of specific results of SBC
testing. We have been fully briefed on the status of the SBC research
program being conducted by the Corps at Lower Granite Dam, and Bonneville
Dam powerhouses. We have also been briefed on feasibility analyses at The
Dalles and John Day dams. In addition, we have been briefed on the results
of 1996 and 1997 prototype tests by Chelan County Public Utility District
(P.U.D.) at Rocky Reach Dam and by Grant County P.U.D. at Wanapum Dam.
Preparatory to this report, the Corps and Chelan P.U.D also briefed the
ISAB on the preliminary results of their studies in the spring of 1998.
In all cases, significant progress has been made in identifying the
relative effects of features of the prototypes. At Lower Granite Dam,
there was an increase in 1998 in the percentage of fish that were diverted
into the surface flow bypass. Tests by the Corps at Bonneville, The Dalles
and John Day dams were of a preliminary nature, but the results are
promising enough to justify larger scale testing. The Dalles Dam in
particular, offers the potential for development of a surface flow bypass
because the ice and trash sluiceway, located above the turbine intakes,
already operates in that mode. It passes 40% or more of the fish
approaching the powerhouse in its present configuration. There is a strong
possibility that its effectiveness can be improved by design
At Rocky Reach Dam in 1998, it was found that an additional upstream
opening intended to attract more fish was not effective. Ideas for a
revised design have resulted. These will be incorporated into prototype
tests scheduled for 1999.
At Wanapum Dam it was found that the prototype surface flow bypass was
not sufficiently effective in collecting fish, but that it was effective
in diverting a high percentage of the fish away from the turbine intakes
and toward the spillway. While no modifications to the prototype were made
for testing in 1998, it was left in place to function in that way, pending
a decision as to its future.
In the same context, surface spill is being investigated as a bypass
alternative. Tests of spill gate configurations that will draw surface
water are being conducted at Rock Island and Wanapum dams and are being
planned for John Day Dam.
Premises for Surface Bypass Development
In order to understand and manage the effects of scientific uncertainty
on implementation of salmon recovery technologies, it is important to
proceed under a guiding set of hypotheses -known as a conceptual
foundation. Given the present configuration and operation of the
hydroelectric system, development of surface bypass is consistent with the
conceptual foundation of the Independent Scientific Group to the extent
that it takes advantage of the natural behavior of the emigrating juvenile
salmon (Coutant 1997). It is important for more specific hypotheses to be
presented and tested regarding how surface bypass technology can improve
the expression of diversity among salmon populations.
More specific hypotheses to guide development of the surface bypass
prototypes have been developed for the U.S. Army Corps of Engineers. The
premises for surface flow bypass (SFB), development, as articulated by
Gary Johnson (Battelle Laboratory) reflect the researchers? present
understanding of fundamental smolt behaviors. The premises are:
- Bulk Flow -- Smolts follow the bulk flow as they migrate
toward the dam. This dictates their lateral and vertical position in
the forebay and predisposes them to pass at certain sites horizontally
along the face of the dam. (Far zone > 150 ft [50 m])
- Opportunity for Discovery -- Smolts are primarily distributed
in the upper portion of the water column, prefer not to sound to pass
dams, and are horizontally distributed such that they are predisposed
to readily pass the dam via a SFB. Smolts can sense the vertical and
horizontal components of the flow regime and use this information to
select migration paths. These behaviors allow smolts the opportunity
to discover the SFB as they move into the near zone. (Intermediate
zone 30-150 ft [10-50 m])
- Entrance Conditions ? Smolts will actively migrate into and
stay within a SFB entrance provided surface flows smoothly accelerate
or decelerate, and acoustical, optical, and biological conditions are
conducive to passage. Entrance conditions include near zone flownets,
total volume, velocity gradients, shape, horizontal/vertical
orientation, light, and surface area. (Near zone: 0-30 ft [0-10 m])
- Conveyance ? Smolts will migrate safely in conveyance
structures, such as channeled flow, dewatering systems, and spillbay-type
- Outfall ? Smolts are not injured in the outfall
environment, and will migrate in the bypass flow as it egresses the
The premises need to be constantly refined and tested against the best
available literature records, as well as against the data developed in the
course of testing.
We stress that the ISAB does not regard these premises as merely
working assumptions. These are all testable hypotheses for which data are
to be collected and evaluations conducted. In future reviews, when these
hypotheses are presented, evidence justifying their retention should also
be presented. If data for retention are not available, alternative
hypotheses should be presented.
We also stress that these premises are not the only applicable
concepts. For example, under the Corps? premise 3, a smooth flow field
is needed in the forebay to attract fish to the collector. We believe an
equally viable alternate premise 3 is that the fish need turbulent
attraction flow, and not a smooth flow field. Development and evaluation
of alternatives is critically important at this point in the process of
In addition to the flow field hypotheses, there are many other
alternative hypotheses and premises that need to be developed. For
example, the five premises of this section presume that there is one
biological entity, the smolt, for which the SFB is applicable. The five
premises could be expanded to include known and inferred behaviors of
multiple life history types within the salmon species, as well as of other
species, such as Pacific lamprey.
Consequently, it is important for the Corps to develop a set of
alternative hypotheses for evaluation. The surface bypass technologies are
so new that having good alternative premises is essential to prevent
spending more money and time than necessary on designing and implementing
features that the fish must have for effective orientation to the surface
- Over 20 years of work to improve turbine intake screen technology
has not yet developed a turbine intake screen that can achieve the 80%
fish passage efficiency (FPE) standard for all species and stocks. It
is time to find an adjunct to the screen technology for bypass.
- Surface collection continues to show promise as a bypass measure
for juvenile salmonids.
- Spill must be provided in an amount to make up the difference
between the 80% FPE standard and the FPE actually achieved by turbine
intake bypass screens.
- The surface collector at Wells Dam is the only bypass system in
the basin that achieves the 80% FPE goal without the addition of
- Surface bypass technology may be able to reduce the amount of
spill needed to achieve the 80% FPE standard.
- The surface bypass technology is not directly transferable from
Wells Dam to other projects because of their different spill way
- Current SBC evaluations speak primarily to the issue of the
potential efficacy of the surface bypass apparatus and not as much to
the specific percentages or numbers of fish that may have been
influenced by its presence.
- Substantial uncertainties remain regarding the level of changes
in survival of juvenile salmon that will be provided by SBC, and
regarding how these changes in survival will be distributed across
listed species, and across the other life history types of the
- Development and application of surface bypass technologies is
necessarily a slow process because of logistical constraints on the
testing of prototypes. The investigators have only a limited number of
alternative configurations that they can test during a given
Arnold, G. P. 1974. Rheotropism in fishes. Biological Review 49:
Coutant, C. C., L. D. Calvin, M. W. Erho, Jr., J. A. Lichatowich, W.
J. Liss, W. E. McConnaha, P. R. Mundy, J. A. Stanford, R. R. Whitney, R.
N. Williams, D. L. Bottom, C. A. Frissell. 1997. The normative river:
an ecological vision for the recovery of Columbia River salmon.
Pages 50-59 in D. J. Mahoney, editor. Waterpower ?97.
Proceedings of the 1997 International Conference on Hydropower. American
Society of Civil Engineers, New York.
Harza 1994. Review of Reservoir Drawdown - Final Report. 1994.
NW Power Planning Council (Portland, OR), Columbia/Snake River Drawdown
Committee. Harza and Associates, Portland OR.
ISAB 98-2. 1998. Response to the Questions of the Implementation
Team Regarding Juvenile Salmon Transportation in the 1998 Season (February
27, 1998). Independent Scientific Advisory Board for the Northwest Power
Planning Council, Portland, OR, National Marine Fisheries Service,
ISAB 98-4: First Report: The ISAB Corps Capital Construction
Project Review. The Scientific Basis for Juvenile Fish Passage
Improvements in the Federal Columbia River Power System: John Day Extended
Length Turbine Intake Screens and Bonneville Dam Bypass System Outfalls
(June 9, 1998). Independent Scientific Advisory Board for the Northwest
Power Planning Council, Portland, OR, National Marine Fisheries Service,
ISAB 98-8. Report of the Independent Scientific Advisory Board
Review of the U.S. Army Corps of Engineers? Capital Construction
Program, II. B. Dissolved Gas Abatement Program. Independent Scientific
Advisory Board for the Northwest Power Planning Council and the National
Marine Fisheries Service.
ISG 96-6. 1996. Return to the River. Report to the Northwest
Power Planning Council, Independent Scientific Group. 584 pp.
Johnson, G.E. 1995. Fisheries research in the forebay of Wells
Dam in spring 1995 related to the surface flow smolt bypass. Pacific
Northwest Laboratory. Report to U.S. Army Corps of Engineers. Contract
No. 19478 Task 12.
Johnson, G.E. and D.D. Dauble 1995. Synthesis of existing
physical and biological information relative to development of a prototype
surface flow bypass system at Lower Granite Dam. Final Report to U.S. Army
Corps of Engineers, Walla Walla, WA.
NOAA 1995. Biological Opinion on 1994-1998 Operation of the
Federal Columbia River Power System and Juvenile Transportation Program.
Reasonable and Prudent Measure 11. March 2, 1995.
Office of Technology Assessment 1995. Fish Passage Technologies:
Protection at Hydropower Facilities. OTA-ENV-641. Washington, D.C.. U.S.
Govt. Printing Office. September, 1995. 167 pp.
Skalski, J. R. 1993. Summary of 3-year bypass efficiency study
at Wells Dam. Processed Report. Public Utility District No. 1 of Douglas
County, East Wenatchee, Washington. 5pp.
SRSRT (Snake River Salmon Recovery Team; Bevan, D.E., P.K.
Bergman, T.C. Bjornn, J.A. Crutchfield, J.P. Harville, P.C. Klingeman,
and J.W. Litchfield) 1993. Draft Snake River salmon recovery plan
recommendations. NMFS/NOAA, Portland, OR. 364 pp.
Whitney, R.R., L.D. Calvin, M.W. Erho, Jr., and C.C. Coutant.
1997. Downstream passage for salmon at hydroelectric projects in the
Columbia River Basin: Development, installation, and evaluation. Northwest
Power Planning Council. Report 97-15. 101 pp.