NITROGEN SUPERSATURATION
IN THE COLUMBIA
AND SNAKE RIVERS
ENVIRONMENTAL
PROTECTION
ABENCY
REGION X
SEATTLE, WASH.
-------
NTTNOCKN SITFMKS ATUI? AT TON IN THE
COLUMHTA AND SNAKIi IlLVliHS
TECHNICAL REPORT NO. TS 09-70-208-016.2
Prepared by:
Robert L, Rulifson, Fishery Biologist
and
George Abel, P. E.
ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF WATER PROGRAMS
SEATTLE, WASHINGTON
REGION X
July, 1971
-------
TABLE OF CONTENTS
Page
I. INTRODUCTION . 1
Purpose 1
Authority 1
Scope 1
Report Organization 3
Fish Species 3
The Physical Environment . 10
II. SUMMARY 12
Conclusions 12
Recommendations 14
III, DESCRIPTION OF THE PROBLEM 18
Physical Evidence 18
Biological Evidence . . . « . .... 21
Chemistry and Physics ........ 26
IV. BIOLOGICAL SOLUTIONS 30
Collection of Downstream Migrants ....... 30
Adjustment of Hatchery Releases 30
Future of Up-River Fish Runs ......... 31
V. ENGINEERING SOLUTIONS ............ 33
Introduction 33
Analysis of the "Base Condition" 39
Proposed Methods of Controlling Nitrogen,
Within Normal Constraints 59
Other Methods of Controlling Nitrogen 79
Effects of Future System Operation ....... 86
Needs for Further Study 98
BIBLIOGRAPHY 100
GLOSSARY 105
APPENDIX
114
-------
ii
LIST OF FIGURES
Figure Page
1 Columbia River Basin Map ^
2 General Timing of Salmonoid Activi-
ties y Columbia River ^
3a Migration of Fish in Upper Columbia
River in Comparison with Nitrogen
Supersaturation . . . . . » . . ®
3b Migration of Fish in Snake River in
Comparison with Nitrogen Super-
saturation ^
3c Migration of Fish in the Columbia River
at Bonneville Dam in Comparison With
Nitrogen Supersaturation ®
3d Migration of Fish in the Columbia
River Estuary iti Comparison with
Nitrogen Supersaturation a.^
4 Temperature - Flow Profiles of Columbia
River , 1967 ^
5 Comparison of Percentage Saturation
of Dissolved Nitrogen in Forebays
from Priest-Rapids Dam to Mouth of
Columbia River, May, July, and August,
1966 19
6 Dissolved Nitrogen Relationships -- Lower
Snake River (Little Goose Project) ^
7 Dissolved Nitrogen Relationships -- Lower
Columbia River 35
B Unregulated Flow at The Dalles, Oregon 38
9 Hydraulic Capacity of Generating Units 41
-------
iii
LIST OF FIGURES (Continued)
Figures Page
10a(l) Base Condition, 1970, High Flow Year .... 42
10a(2) Base Condition, 1970, Average Flow
Year (1956) 44
10a(3) Base Condition, 1970, Low Flow Year 46
10b(l) Base Condition, 1980, High Flow Year (1956) . 48
10b(2) Base Condition, 1980, Average Flow Year
(Columbia - 1952, Snake - 1932) ..... . 50
10b(3) Base Condition, 1980, Low Flow Year (1949) . 52
11. Annual Loads and Resources, Pacific
Northwest ...... 55
12. Hydrographs of Flow at The Dalles Resulting
from Modified Operation of Grand Coulee
Dam (1933 Flood) 61
13. Basic Layout of Structures in a Typical
Hydroelectric Project (Rocky Reach) .... 68
14. Section Through Skeleton Bay -- Little
Goose Project ..... 69
15. Section Through a Typical Turbine - Generator
Bay 69
16. Prototype Slotted Gate Being Installed at
Little Goose Project. ........... 70
17. Total Potential Diversion Capacity -- Columbia 76
and Snake River Projects
18. Proposed Configuration of a Low-Level
Regulating Outlet (Lower Granite Spillway) 78
19. Proposed Configuration of Spillway Deflector
(Lower Granite Spillway)
-------
iv
LIST OF FIGURE^ (Continued)
Figures Page
20a Effect of Additional Upstream Storage,
High Flow Year (1956)
20b Effect of Additional Upstream Storage,
Average Flow Year (Colombia - 1952,
Snake - 1932)
21 Daily Load Shape, Federal System . e » . ° 88
22a Load Factoring at Hydro Plants Only ... 97
22b Load Factoring at Thermal Plants and
Hydro Plants 97
-------
V
LIST OF TABLES
Table Page
1 Average Annual Upstream Migrant Fish
Counts at Bonneville Dam By Species ... 4
2 Estimated Costs for Installation of
Slotted Gates in "Lower Snake River
Projects 72
3 Future Electric Power Requirements,
Columbia-North Pacific Region 89
4 Installation Schedule for Thermal
Projects ^
5 Installation Schedule for Hydro Projects . . 92
6 Load Resource Analysis: Plan A,
Columbia-North Pacific Region 96
-------
ACKN OWI.TCDG KMENT
vi
The authors would like to express their appreciation for the
valuable assistance received from personnel of several State and
Federal agencies who cooperated in the preparation of this report.
Special acknowledgement Is given to Hugh Smith, Doug Speers, and
Nick Dodge of the U. S, Army Corps of Engineers, North Pacific
Division; Don Baldrica of Bonneville Power Administration, U. S.
Department of Interior; Carl Elling, National Marine Fisheries
Service; Kirk Heinengen and Charles Junge, Fish Commission of
Oregon; Daniel Krawczyk, Environmental Protection Agency; and Dr.
Raymond Weiss, Scripps Institution of Oceanography.
-------
I - INTRODUCTION
Purpose
The purpose of this report is to describe the problem of
dissolved nitrogen gas supersaturation including its effect on
fish in the Columbia River and recommend solutions to the prob-
lem, This information is the status of present knowledge of the
causes and effects of the supersaturation and is the basis for
conclusions and recommendations for short and long range solutions
on the Columbia and Snake Rivers. The same principles would apply
to rivers in other areas as well.
Authority
Nitrogen in supersaturation has been shorn to be a toxicant
which causes mortalities to aquatic life and as such violates the
criteria of the Interstate Water Quality Standards pursuant to
the Water Quality Act of 1965,
In accordance with the responsibility of the Environmental
Protection Agency (EPA) as set forth in Executive Order 11507 of
February A, 1970, Mr. William D, Ruckelshaus, Administrator,
directed the Northwest Regional Office of the Environmental Pro-
tection Agency to prepare a report outlining the scope of the
nitrogen supersaturation problem in the Columbia Basin with
recommendations for its solution.
The information and data presented in this report are from
studies and observations made on the Columbia and Snake Rivers.
Figure 1 presents a map of the Columbia Basin. Some of the
material presented is from published information but much of
it is from progress reports and personal communications with
representatives of the northwest state and federal fisheries
agencies, the U. S. Army Corps of Engineers, North Pacific Div-
ision (NPD), and Bonneville Power Administration (BPA). While
solutions are being sought to the problem of dissolved nitrogen
supersaturation specifically on the Columbia River, the results
of the studies are applicable to any location where water is
spilled at dams or in some cases at natural falls.
-------
2
RIVER BASIN
(TTrmr—— existing ma.ior pwojccts
( .. MAJOR RC^CRVOIR PROJECTS
— " UNDER CONSTRUCTION
Anadromous Fish
Spawning Area
Figure 1. Columbia River Basin Map
-------
3
Report Organization
Chapter II presents a summary of the major conclusions and
recommendations reached from the technical evidence. Chapter III
gives a definition of the problem based upon the physical, chemi-
cal and biological evidence. Biological solutions or modifica-
tions arc discussed in Chapter IV. A description and analysis of
engineering solutions to the problem are described in Chapter V.
Fish Species
The Columbia River is inhabited by four species of Pacific
salmon: sockeye salmon, chum salmon, coho salmon, and chinook
salmon. The chinook salmon has distinct spring, summer, and fall
races, while the other species do not0 Another significant
anadromous number of the salmonid family is the steelhead trout,
which has both summer and winter races. Other anadromous fish
of the Columbia Basin include several other species of trout,
smelt, and American sliad. Non-migrating fish species include
the white sturgeon and several species of spiny rays (bass,
crappie) as well as numerous other forage fish and predators.
Hie anadromous species seek out Columbia system waters
far inland from the Pacific Ocean, historically reaching as
far as Canada. They may presently be found in nearly all acces-
sible tributaries where water quality conditions allow. Figure 1
shows the areas which remain accessible to anadromous fish since
construction of the system of dams and reservoirs in the basin.
Spawning occurs in the tributaries and in the only remaining unim-
pounded waters of the main-stem Columbia, the fifty-mile reach
below Priest Rapids dam'and through the Atomic Energy Commission's
Hanford Reservation, which produces large numbers of chinook
salmon.
The size of the Columbia River fish resource is most easily
described in terms of the fish count over Bonneville Dam. In
addition, it should be remembered that the Willamette River,
Cowlitz River and smaller tributaries produce significant addi-
tional numbers of fish, and that the commercial and sports catch
of the fish in these rivers and the Pacific Ocean from Alaska to
California should be added to the Bonneville count. The thirty-
one year record of upstream migrating fish passing over Bonneville
Dam gives an annual average count of 705,875 salmon of all species
and steelhead. The populations of the various species can be
summarized as shown in Table I. Hie total numbers of fish
entering the Columbia River in recent years is considered
-------
4
by fisheries agencies to be as high as in the 1930's. However,
the size of the fish resource is now largely dependent upon
artificial production. Also, the distribution of species and
races making up the total has been changed by man's alteration
of the river system.
TABLE I
AVERAGE ANNUAL UPSTREAM MIGRANT FISH COUNTS
AT BONNEVILLE DAM BY SPECIES
1938-1968
Chinook salmon .................. 385,252
Sockcye salmon 100,777
Coho salmon (Three-year average 1965-1968) .... 77,289
Chum salmon 1,214
Steelhead trout . 141,343
Source: Annual Fish Passage Report. Columbia River Project, U.S.
Army, Corps of Engineers, North Pacific Division,
Portland, Oregon.
All of the salmonids have similar life histories or cycles
but each species and race matures at different rates, presenting
differences in the duration and time of occurrence for the va-ious
stages of activity: spawning and incubation, rearing, juvenile
migration, growth at sea, and adult migration. For example, sal-
mon fry spend from a few weeks to over a year in fresh water
before beginning seaward migration, depending upon the species
and race. Similarly, the amount of timj/spent at sea also varies
from less than one year- for jack salmon— to over five years,
depending upon the species and race of fish. Figure 2 illustrates
the time of occurrence during, the year of the various stages of
freshwater activity for the chinook, coho, and sockeyp salmon,
and for steelhead trout. As can be seen, migrating adults or
juveniles can be found in the Columbia River throughout the en-
tire year. However, the peak of migration occurs in a much
shorter time. This is shown in Figures 3a, 3b, 3c and 3d, for
four locations relative to the nitrogen concentration.
1/ Jack salmon are sexually precocious males.
-------
AD'JLT MIC RATION
" zizziiz!
CHINOOK
ZZl = pcriO'J cf C'.'w.'.y
spav.t::::g h i:;cusat:o:'|
REARING
:-;>v- ^; I M
JUVENILE MIGRATION
-j
. • j
. * l
ii j
COHO
ADULT MIC RATION
1 I I I !
^AV.'NIVG
-------
6
Figure 3a. Migration of Fish in Upper Columbia
River in Comparison with Nitrogen
Supersaturation
(Prepared by National Marine Fisheries Service)
3= 120
Adults
Ll.
iL
o
in GO
o
% 60
3
O ^
x 40
UPPER COLUMBIA RIVER (above Snake)
7/7\ CHINOOK "
f3.STEEl.HDL
SOCKEYE'
CO HO
SHAD "
MJJASOND
MONTH
Juveniles
[23 CHINOOK (Is)
D7TTI .-I .
\////f\ ^Ub)
STEELMEAD
COHO
SHAD
SOCICEYE
O
140 l-
-------
7
120
100
00
GO
40
20
0
15
SNAKE RIVER
Adults
YZ3 CHINOOK
J&J STEEL HEAD
lUll SOCKEYE
E3 C0H0
E
SHAD
W
¦m
E-'fzS
p7Vsff S/\///)//.# •>
Juveniles
¦^.v. »•&¦:- y AtA
CHINOOK (Is)
3 CHINOOK (O's)
r- 3 rv-T- ir- «r- «
^ ' LCLiiLAU
10
0
140 f~
120 DC
Ld
0.
nn 3
2
100 LU
o
O
oc
h-
M A
J J A
MONTH
0 N
D
Figure 3b, Migration of Fish in Snake River
in Comparison with Nitrogen
Supersaturation
(Prepared by National Marine Fisheries Service)
-------
8
X
CO
\Z
Ls_
o
co
Q
<
CO
Z>
O
X
X
rn
U.
Ll.
O
CO
2:
O
600-
500
400
300
200
100
0
30
2b
20
BONNEVILLE DAM
Adults
YZ/\ CHINOOK
£~'] STEEL HEAD
FBI] SOCKEYE
w C0H0
[ml SHAD
5
10
5
0
!. ,/ ,, ~
y - ¦'V.' >/S/
Juveniles
CHINOOK (Is)
CHINOOK (0's)
r.""! STEELHEAD
sa C0H0
f=st§1 SHAD
[~rj SOCKEYE
iwl k'^u„/jT77777\ t-_-—-—> ^-*1==: ~—1-
ezzzzzk*-•¦ V-^ < ?TT7-,rmr^72J2n
M A M J J
140 g
or
130 h
<
CO
120 k
LaJ
-------
9
600j-
500
co 400
Q
g 300r-
3
O
x 200
h-
oc
0
ESTUARY
Adults
\Y/\ CHINOOK ~
STEELHQ.
gM ?li-soc:cEVE-
C0l"i0
SHAD
. h
• — . •—j.-is/\/Vv s'V
...... "... *. • y^V\'¦ . \/\' \ • i
'vAA/y V V\ 5
• - ~ vvyv.' ^vj
.. nA'W N/N'V?
—----- VWS'-'V^V:
30
r 25
U-
u_
O 20
2 I 5
O ' J
_]
=J 10
5
0
yy.
- "•.... \/\ / V ... A/\ I
. • \/VN V- ' I
p:=sr - ¦ : ^-x v:v.)0)
|—v:- J. ' ~y \
W$WA y/o\
•• '^Z/Ov/
Wk^ZL^/07yi.
/ »r / / //l , I
Juveniles
CHINOOK it's)
m}x CHirvOc:-: (o's)
ifj STEELHE-
>3 C0H0
ij SHAD
# SOCUEYE
140 H
DC
I oO f-
<
CO
120 ^
UJ
a.
100 u
o
o
cr
h-
J A S
MONTH
Figure 3d. Migration of Fish in the Columbia
River Estuary in Comparison With
Nitrogen Supersaturation
(Prepared by National Marine Fisheries Service)
-------
10
The Physical Environment
The Columbia River remains accessible to anadromous fish as
far upstream as Chief Joseph Dam, river mile (RM) 545. Throughout
this length, two reaches are unimpounded: the estuary to Bonne-
ville Dam (RM 145) and the llanford reach between the head of
McNary Pool (RM 347) and Priest Rapids Dam (RM 397)o The remainder
of the river is impounded behind nine run-of-the-river dams, with
shallow reservoirs ranging from 54 to 165 feet in depth (Figure 1)„
The surface runoff in the Columbia Basin is characterized by
a typical snowmelt regime, with low flows during the late summer
and winter and high flows during the spring and early summer,,
The volume of runoff in the Columbia ranks it as the fourth largest
in the North American Continent; the average annual runoff at The
Dalles, Oregon is 140 million acre-feet. Mean annual flow is
195,400 cubic feet per second (cfs); mean monthly discharges range
from 95,700 cfs in January to 494,700 cfs in June.
With minor exceptions, the water quality of the Columbia
River—other than temperature and dissolved nitrogen—is not con-
sidered detrimental to the fish resources. Water temperatures
in the main stem Columbia River to Priest Rapids Dam range from
maximums in the 70's during warm summer months to lows in the
40's and occasionally in the 30's during winter. Figure 4 shows
water temperature and discharge profiles for various points on
the Columbia River to illustrate general temperature and flow
conditions.
-------
ID
u.
u
«n
O
O
O.
LlI
cr
<
X
l_i
in
Q
<
LU
2
5
X
f-
z
o
to
LU
UJ
K
en
LU
O
Z
lL)
g:
o
£
d:
ID
Ql
S
iu
»-
00
MAX. MONTHUf MEA'I TEMR
Figure 4. Temperature-Flow Profiles of
Columbia River, 1967
-------
II - S1JMMAKY
Conclusions
1. Essentially all reaches of the Columbia and Snake Rivers down-
stream from hydroelectric dams are significantly supersaturated with
dissolved atmospheric gasses during periods of high spillway discharge.
2. The salmon and steelhead runs in the Columbia and Snake
Rivers are seriously jeopardized under present conditions by super-
saturation of dissolved gasses.
3. Nitrogen supersaturation conditions in the Columbia and
Snake Rivers are caused at main stem hydroelectric dams when river
flow must be passed over the spillways during the spring flood
period. Nitrogen in the air is entrained in the flow as it passes
over the spillway and, as the water plunges into the stilling basin
below the dam, increased pressure forces the nitrogen from a gaseous
state into solution in the water,
4. Water passing through the power-generating turbines or
otherwise not exposed to the aerating effect of the spillways does
not become supersaturated with nitrogen.
5. Nitrogen supersaturation levels above 105 percent produce
symptoms of gas bubble disease in fish, and levels above 120 percent
are lethal.
6. Nitrogen, because of its greater proportionate volume in
air, has been measured and related to toxic effects on fish. How-
ever, it is known that the other atmospheric gasses are also in-
volved in the gas bubble disease phenomenon.
7. Hypothetical flow regulation stadies show that, with no
operational or structural modifications for control of nitrogen,
spillway flows at existing dams will be great enough to produce
lethal levels of nitrogen supersaturation in years of average and
higher river flows but not in low flow years. An average year is one
with flows which, statistically, would be equalled or exceeded one
year out of two. Although it is not certain at what specific river
flow levels nitrogen problems begin to occur, it is apparent that the
threshold level is somewhere between low and average flows.
8. Under present plans to expand the Columbia Basin hydroelectric
system through the year 1980, the volume of spills at the various pro-
jects will be reduced. However, without additional control measures,
the reduction in the volume of spills will not be great enough to re-
duce nitrogen supersaturation to levels considered safe for fish during
even a year of average flows.
-------
13
9. Installation of slotted gates and bypass of flow
through empty turbine bays in the Lower Snake River dams will
further reduce the spills in that river, but nitrogen levels
hazardous to fish will still occur in years of average or
higher flows.
10, A basic requirement to a solution of the problem is
to devise methods whereby volume in excess of flows through
turbines can be passed by dams without entraining atmospheric
gasses.
1], Lower Granite Dam, presently under construction,
will add to already critical nitrogen problems in the Lower
Snake River unless the completed project includes provisions
for passing flood flows without entraining atmospheric gasses.
12. Unless provisions can be made to pass maximum flows
through turbines, installation of turbine generator units in
dams before they are needed to meet regional power demands may
eliminate flexibility in the use of the otherwise empty turbine
bays to bypass flow, and thus may aggravate the nitrogen
problem.
13. Increasing use of the hydroelectric system for peak
power generation could cause wide diurnal fluctuations in the
amount of spilling unless pumped storage, pondage, and thermal
power facilities are effectively used to reduce the fluctuations.
14. The critical nature of the problem in 1971 caused an
unprecedented level of coordination and cooperation among the
Corps of Engineers, Bonneville Power Administration, and the
Northwest fishery agencies in developing partial immediate solu-
tions to the problem,'
15. Flexibility within the federal dam-power complex con-
tributed much to the success of a flow reduction program in
April 1971 to make the best conditions for a mass hatchery fish
release. Although the program required increased coordination,
the cost to the Government was minimal. Continued annual use
of the flexibility of the federal system is a necessary partial
solution to the nitrogen problem.
16. The Washington Public Utility Districts and Idaho Power
Company cooperated in the nitrogen reduction program in 1971, but
-------
14
even ^renter interchange botwecn the federal and noti-fcdcral
systcnis should be permitted by necessary changes in regulations
and licenses. It is necessary to include the Federal Power
Commission in future planning.
17. Trapping of downstream migrants from the Snake River
and transport to the estuary are a necessary partial solu-tion
to the problem. Additional trapping at McNary Dam would salvage
fish from the Upper Columbia and residual from the Snake River.
18. Delay in passage of adult upstream migrants below
dams and in fishways is a significant factor in increasing
mortalities from direct or secondary effects of gas bubble
disease. Changes in the water management system within allow-
able flexibility could relieve the adult fish delay problem.
19. Nitrogen conditions are more severe at some dams than
at others because some necessitate more spilling than others.
Also, spillway and stilling basin designs have an effect.
Dams newly completed have required almost complete spilling of
flood flows until the powef-generating turbines are installed --
for example, at^John Day Dam, completed in 1968.
Recommendations
1, The states of Oregon, Washington, and Idaho, together
with the Environmental Protection Agency, should initiate imme-
diate action to establish a water quality standards criterion
for dissolved nitrogen in the Columbia and Snake Rivers.
a. The standards criterion should establish a maxi-
mum allowable concentration of dissolved nitro-
gen in the Columbia and Snake Rivers at 110 percent
of saturation based on analytical procedures
presently being followed by the National Marine
Fisheries Service.
b. Immediate research and development efforts should
be initiated to develop a method to measure
total dissolved gas partial pressures to be
related to dissolved gas concentration data.
-------
15
c, Expanded research and development efforts are also
needed to relate dissolved gas partial pressure
data to effects on fish. The results of these
studies may be used for future reviews of the
standards criterion for nitrogen.
d. The Environmental Protection Agency should work
with the National Oceanic and Atmospheric Admin-
istration to develop an effective regional program
for dissolved nitrogen monitoring in the Columbia
and Snake Rivers, A formalized monitoring pro-
gram will be needed to measure progress towards
compliance with the standards criteria,
2, A regional council should be established iTimediately
to make specific recommendations on administrative, legislative,
and policy actions required to implement an effective nitrogen
control program. Membership on the council should include "high-
level representatives of regional environmental, fisheries,
water resource development, and water management agencies. The
regional council for nitrogen control should periodically report
on its review of the nitrogen supersaturation problem, assess-
ment of progress towards control of nitrogen supersaturation
effects on fish, and recommendations for further action to the
council's constituent agencies, the Council cti Environmental
Quality, and other affected agencies. Elements of a regional
nitrogen control program should include the following:
a. Nitrogen supersaturation reduction program,
(1) Maximum utilization of the flexibility of the
regional system pf reservoirs and power-
generating facilities to reduce spills during
periods of fish migration,
(2) Provision at each dam of a method .to pass
any flows which are in excess of those
required for power generation through or
over facilities which will not entrain
excessive atmospheric gasses in the water.
The Corps of Engineers should be fully sup-
ported in their suggested program of
structural modifications for this purpose
at their projects and others.
-------
16
(3) Revisions of generator installation sched-
ules and construction methods to maintain
maximum bypass capability at dams during
the period of construction. Congressional
authorization and appropriation schedules
may need to be adjusted appropriately.
b. Expanded program to physically limit exposure of
migrating juvenile saltnonids to nitrogen super-
saturated waters.
(1) Expand the installation and operation of
fish collecting screens in all turbine
units at Little Goose Dam by April, 1972.
The collection screens could later be
transferred to Lower Granite Dam at the time
of its completion. Lower Granite should be
designed to allow the installation and
operation of both skeleton bays and fish
collection screens.
(2) Install and operate fish collection screens
in all turbine units at McNary Dam by
April, 1973.
(3) Continue to expand the program initiated in
1971 to make mass fish hatchery releases
during periods of reduced nitrogen. Trans-
port fish, where feasible, for release in
the Lower Columbia River below Bonneville
Dam,
3. Any new dani in the Columbia Basin must include facili-
ties for adequate control of nitrogen supersaturation. Speci-
fically, adequate controls at Lower Granite Dam (now under
construction) must be provided before completion of this project.
The design goal for these control facilities should be to maintain
dissolved nitrogen concentrations downstream from the project
below 110 percent of saturation in a ten-year flood.
Additionally, existing plans for further alteration or
expansion of the Columbia River water resource development system,
including hydroelectric power-peaking aspects of the hydro-thermal
-------
17
program, should bp reviewed by tlie river management agencies
to insure that the plans are consistent with the above strategy
for abatement of nitrogen supersaturation problems. In order
that control strategies can be developed and implemented on
a timely basis, this review should be completed and provided
to the regional nitrogen control council by January, 19720
-------
Ill - DESCRIPTION OF T11E PROBLEM
The Washington Department of Fisheries found in the early
1960's that levels of dissolved nitrogen (N2 ) supersaturation
in the Columbia River were sufficiently high to cause problems
with salmon runs. Previous experiei.^e with nitrogen supersatura-
tion related the problem principally to hatchery water supplies
(Rucker & Hodgeboom, 1953). Following this discovery, a con-
clusion was reached by the technical staffs of the Northwest
fishery agencies that if nitrogen was an important cause of
fish mortalities, the problem x^ould eventually be eliminated with
the completion of upriver dams which would eliminate most of the
spill. As more information was assembled, this conclusion proved
to be erroneous.
This problem was recognized by the Environmental Protection
Agency as one of great magnitude which threatens the continued
existence of runs of valuable salmon and steelhead on the Columbia
River. The Washington,State Department of Ecology has likewise
recognized this problem and has asked assistance of EPA and the
Corps of Engineers in its solution.
Physical Evidence
Dissolved nitrogen saturation levels as high as 125% were
observed in the spring of 1965 at some sites in the Columbia
River by the Washington Department of Fisheries and the Bureau
of Commercial Fisheries. The adverse effects of saturation levels
of 125% are well documented. Therefore, additional surveys were
made in 1966-67 to determine the cause of. supersaturation and
seasonal variations and to determine the effect on adult and
juvenile salmon (Ebel, 1970C). Water samples were collected in
1966 from 26 stations between Grand Coulee Dam forebay and the
estuary at At toria. The samples were analyzed for dissolved
nitrogen and the results of four series are shown in Figure 5.
The evidence shows that supersaturation in the Columbia River
is caused primarily by spillways at dams. During periods of the
year when spilling normally does not occur, the saturation level
is near normal (100%) but increases to levels as high as 140%
during periods of spill. To further test the effect of spilling
on saturation values, a series of tests was run at Bonneville
Dam in March 1966, a period when normally ther^. would be no
-------
7U
O
H
H
c
cj
o
M
H
C
H
in
$
a£
o
H
CO
<
&
W
M
>
o
s
2
<
p
O
on
~3
w
J
iJ
H
iJ
>
<
W
a
O
s
pq
H
W
H
M
p
C/J
a
H
2
H
£3
D
Q
¦<
Q
£
O
2
<
e
p
D
pa
<
Q
>i
g
cJ
<1
8
c
M
O
o
ss
C
S
CO
>*
_ _ _ - 1966
1968
1967
1968
- 1966
1967
1968
1966
100
200 300 400 500
KILOMETERS FROM MOUTH OF RIVER
600
Figure 5. Comparison of Percentage Saturation of
Dissolved Nitrogen in Forebays from
Priest Rapids Dam to Mouth of Columbia
River, May, July, and August.
1966 (short dashes) , 1967 (long dashes) ,
1968 (solid line).
(Beitiingen and Ebel, 1971)
-------
20
.spill (Kbel, op. c i L".). The results show that "during the spill
tesLs, conelnl r.iI ions below the spill increased to 1257<> satura-
tion, whereau concentrations in the forebay and below turbines
remained over 100%."
Subsequent studies in 1967 and 1968 confirm the general
findings of 1966, that during periods of spill the nitrogen
levels increase. These data are included in Figure 5. One
observation of importance is that equilibration of the dis-
solved nitrogen concentration in the reservoirs between dams
is very limited, probably because of lack of circulation in
surface waters. An exception has occurred in John Day Reser-
voir, possibly because of its greater length. A greater amount
of water passing over the spillway at John Day in 1968, the
year of initial operation, caused a drastic increase in nitro-
gen in its tailracc. Peak nitrogen concentrations ranged
from 123 to 1457-. The effect of ihis increase on The Dalles
Reservoir downsLrcam is shown on Figure 5. With more water
passing through the turbines and less over the spillway at
John Day Dam in 1969-70, the effect is less severe. Surveil-
lance of the Columbia'and lower Snake River has continued in
1970 and 1971 by the National Marine Fisheries Service (NMFS),
(formerly Bureau of Commercial Fisheries) funded by the Corps
of Engineers, Results confirm those of previous years, that
during periods of excess spill, dissolved nitrogen levels in-
crease below the uppermost dam (Grand Coulee) and remain high
to the mouth of the Columbia. The runoff started earlier in
1971 than in prior years and by mid-April the nitrogen concen-
tration had reached dangerous levels. Data collected by the
NMFS for 1964-69 has been published by Beiningen & Ebel (1971).
The EPA has sampled for nitrogen upstream and downstream
from Oxbow and Hells Canyon Dams on the "Snake River in 1969,
1970, and 1971. The results of analysis show either a sub-
saturation or slight supersaturation in Brownlee Reservoir, but
a moderate to high increase in the tailrace and in the stream
reach below depending on the volume of spill at the upstream
dams.
Two sampling runs in April and May, 1971, from river
mile 239 below Hells Canyon Dam downstream to Lewis ton (RM 141)
and Little Goose Reservoir (RM 72), respectively, give a-picture
of variable nitrogen concentrations, with a general decrease
downstream indicating slow equilibration. Levels from 120-133%
were observed below Hells Canyon Dam decreasing to near 1107o
at Lewiston and increasing again in Little Goose Reservoir.
-------
21
The reason for the variability cannot be explained at this
time. Data from nnjor tributaries (Grand Rondo and Ininaha
Rivers) near their uioulhs, show citlicr equilibration or low
suparsaturntion. The Salmon River, on the other hand, con-
tained nitrogen from 110-114% and the Clearwater River showed
1127c saturation.
Washington Department of Fisheries reported what they
considered to be wide variability of nitrogen saturation in
the forcbay of Chief Joseph Dam, (Heekin, 1971). Harvey and
Cooper (1%2) found considerable variation in oxygen satura-
tion during one day and with the level of discharge constant
below a falls on a natural stream. They thought changes in
barometric pressure and temperature were responsible.
Biological Evidence
Observations of Dead Fish and Fish Showing Symptoms of Gas
Bubble Disease
A number of observations of dead and affected adult and
juvenile salmon and steelhead have been observed and recorded
on the Columbia River in recent years. While some of the
evidence is circumstantial, the number of observations and
circumstances leave little doubt that dissolved nitrogen super-
saturation is responsible. The best summary of these observa-
tions is contained in an address to the Tri-State Governors
meeting on March 23, 1971 (Schoning, 1971). The most convincing
evidence of the severity of the problem is the results of the
juvenile salmon marking study on the Snake River where mor-
talities of downstream migrants were less than 57o prior to com-
pletion of Lower Monumental and Little Goose Dams; after
completion in 1970, the mortality to naturally reared marked
fish was estimated to be 70% and largely attributed to nitrogen
caused problems. Mortality to hatchery reared chinook has
been demonstrated to be even higher (Raymond, 1970B).
External symptoms of gas-bubble disease include: eyes
distended and bcmorrUnged; bubbles in the subcutaneous layers
surrounding the eyeball; enlarged vesicles between the epi-
thelium and connective tissue in the roof of the mouth and
on the outside of the operculum (Beiningen and Ebel, 1970);
(Bouck, Chapman, Schneider, and Stephens, 197CB)0
Large numbers of adult salmon bearing characteristic
symptoms of gas-bubble disease have been observed on several
-------
22
occasions. The most serious incident: occurred at John Day
Dam in 1968, the first year of operation of the fish facili-
ties when no turbine facilities were operational and all dis-
charge was by way of the spillway. Three separate delays
were recorded during the 1968 migration season, two at John
Day and one at The Dalles Dam. The highest count of dead
fish observed after a delay at John Day Dam was 13 sockeye
and 365 chinook on July 29. Fish Commission of Oregon esti-
mates place the loss between Bonneville and McNary Dams at
over 20,000 summer chinook salmon. High nitrogen concentra-
tions were implicated in this loss.
In the opinion of liouck et al. (1970A), the immediate
mortality may noC be the most serious consequence of super-
saturation to adult salmon. In connection with a thermal
effects study on adult salmon, Bouck et al. (op. cit.) removed
adult sockeye salmon from the Columbia River during a period
of high nitrogen supersaturation and subjected them to air
equilibrated water with temperatures ranging between 50 and
70 F (10 - 22.2 C). Loss of eyes or blindness which was at-
tributed to gas-bubble disease occurred in 1/3 to 2/3 of the
fish. The sublethal damage such as blindness may be essen-
tially equivalent to death in terms of damage to natural
reproductive potential (Westgard, 1964; Bouck et al., op,
cit.). The disease may have been responsible for a 51% below
average number of spring chinook reaching the Snake River
spawning grounds in 1968 in spite of a record high number of
adults passing from the Columbia into the Snake. Coutant &
Genoway (1968), concluded that the harmful effects of high
water temperatures on adult salmon are worsened by prior or
simultaneous exposure to nitrogen supersaturation. The sur-
vival time of jack chinook acclimated to 62.2 F (17C) which
were subjected to 71.6 F water (22C) was drastically shortened
when the warmer water contained dissolved nitrogen gas levels
in excess of 11570 of saturation. Regardless of the nitrogen
levels in the warmer water, acclimatization of the fish to
water supersaturated with nitrogen reduced subsequent survival
time.
Delay in Passage
Studies by the Fish Commission of Oregon at The Dalles
and Priest Rapids Dams show a significant delay in passage of
adult salmon and steelhead on weekdays as compared to weekends
(Charles Junge, personal communication). The delay is related
to the effect of the peaking type of power generation. It
seems reasonable to assume that any delay of fish below a dam
or in a fishway during a period of high nitrogen conditions
will increase their exposure. This would be particularly
-------
23
harmful if the delay was in a fishway where the fish are forced
to stay in relatively shallow water.
Live Cape TesLs
The question of the compensating factor of depth when
nitrogen supersaturation is present has been explored by the
National Marine Fisheries Service in experiments at Priest
Rapids Dam, 1967, and at Ice Harbor Dam, 1970 (Ebel, 1970B;
Ebel, 1970C), The results show generally that fingerling
salmon held at the surface or in cages at depths up to 7 feet
suffer excessive mortalities approaching 100% while fish in
cages at about 10 feet depth do not suffer mortalities due
to the protection of increased hydrostatic pressure. In a
cage where the fish were allowed to sound at their own voli-
tion to 15 feet, the mortality was reduced but still ranged
from 40 to 68 percent after 7 days exposure. This test may
approach actual conditions the fish may experience in the
river and also indicates the fish may be unable to detect the
nitrogen and thus avoid it.
Problems in Other Areas
The problem of nitrogen supersaturation is hol restricted
to the Columbia and Snake Rivers. It has been reported by the
Fish Commission of Oregon as responsible for a fish kill at a
salmon rearing station (Dexter) during a period of high spill
at an upstream dam. Two other instances of extensive fish
kills were the result of air being drawn into a pipeline (Wyatt,
1969). Harvey & Cooper (1962) reported the problem from a
Canadian stream used for a hatchery water supply. A falls
upstream caused the supersaturation. Others have reported the
problem associated with hatchery operations (Rucker & Hodgeboom,
1953; Rucker & Tuttle, 1948; Westgard, 1964; Harvey & Smith,
1961; Marsh & Gorham, 1904).
Safe Level of Nitrogen
A number of experiments beginning with Marsh & Gorham
(1904) have been conducted to determine safe levels of dis-
solved gasses for hatchery use. Nitrogen is the usual cause
of gas embolism because it is the least soluble of the three
principal gasses in air and is not removed from the dissolved
state by chemical combination with blood components. Compar-
ison of results and methods of different investigators is
difficult. However, two researchers agree that nitrogen satura-
tion of less than 1107o causes symptoms of gas-bubble disease
-------
24
or ck\ith to salmon ids (H.irvoy and Cooper, 196?.; Shirahata,
1966). OLher:; report levels between 1107„ and 120% as not
satisfactory (Rueker and Ti ttlc, 1948; Sucker and Ilodgeboon,
1953)„ Researchers arc agreed that dissolved nitrogen concen-
trations above l207o of saturation are lethal,, The sublethal
effects of chronic exposure to the lower concentrations is not
well understood and requires additional research.
Pathology
A number of investigators have described the gross symptoms
of gas-bubble disease: bubbles appearing under the skin and in
the fins and eyes of fish, but Pauley and Nakatani (1967), were
the first to describe the histopathology. Tissues from many organs
and parts of chinook salmon fingerlings that exhibited a light
but detectable case of gas-bubble disease from external symptoms
were examined histologically. All tissues of the diseased fish
except the hearts and stomachs exhibited definite histopathologi-
cal changes with rupture of tissue cells and red blood cells in
the gills and kidneys. Tissues were swollen and filled with fluid.
They particularly noted the swollen condition in the roof of
the mouth which seems to be characteristic of this disease.
Other investigators have noted similar changes with the exception
of the roof of the mouth due to chemical toxicity. All investi-
gators are Ln agreement that death from acute exposure is caused
by blockage of vital organs and tissues by gas emboli. An em-
bolism reaching the heart causes instantaneous death. One
aspect of the disease that pu2zled Pauley and Nakatani (1967)
and other investigators is the apparent recovery of diseased
fish when removed from the high nitrogen environment,,
While no work hat, been done to prove the point, pathologists
are in agreement that microbubbles known to be present in turbu-
lent ly mixed water probably have no part in the pathology of
gas-bubble disease and only dissolved nitrogen is involved.
Shirahata (1966) determined the composition of gas which accumu-
lated in abdomens of his test fish and found it similar to that
of gas bubbles in rearing water and of atmospheric gas. He
believes the gas in abdomens may originate from the free gas
bubbles in water. If this is the case, the gas transfer cannot
take place directly but must pass scross two membranes.
One conclusion that can be reached from the limited informa-
tion available is that the etiology of gas-bubble disease is
incompletely understood and is in need of further research, par-
ticularly to learn the effects of chronic exposure to low levels
of supersaturated gasses.
-------
25
Embolism
It is agreed that gas bubble disease is caused by the forma-
tion of £os emboli (bubbles) within the fish. The blood dis-
solved £<,as level closely follows that of the surrounding water.
If water is supersaturated, the greater partial pressure of
atmospheric gasses jn the water will cause the blood and tissues
to also became supersaturated. One idea that has been suggested
for the formation of bubbles within blood is because of the
turbulence of flow or pressure drop within the circulatory system.
That this could occur in fish seems unlikely considering the low
hydrostatic pressure under which most fish circulatory systems
operate, A hypothesis more reasonable than the blood turbulence
idea is presented by a Battellc-Northwest physiologist (Hark
Schneider, personal communication). He believes that blood is
an unstable liquid offering many sites for emboli formation in-
cluding protein and lipid molecules, red blood cell membranes
and gas micronuclei. Hill's (1967) showed that fluids super-
saturated ultJi gas are inherently unstable and uill revert to
a stable thermodynamic state with the release of free gas bubbles
if given a site for initial bubble formation. The gas micronuclei
may always be present in blood but are so small that they them-
selves never cause embolism and only become a problem when the
blood is supersaturated and therefore unstable. The question of
gas micronuclei was extensively considered by Harvey (1951).
He had previously said the origin of gas micronuclei was obscure
but an important consideration (Harvey et al., 1944). Evans &
Walder (1969) demonstrated in their work with shrimp that micro-
bubbles are involved in formation of larger internal gas bubbles
upon decompression of the animal. They also showed that forma-
tion of new microbubbles is associated with muscular activity.
In Schneider's hypothesis, he assumes that with temperature and
pressure constant there is a threshold level of blood dissolved
gas below which no emboli will form.
The fish gas bubble disease has been described as closely
related to the bends experienced by human divers following rapid
decompression. Technically, the "bends" (Caisson disease) is a
condition of embolism in a human diver caused by too rapid
decompression of air. Distinction should be made that embolism
in fish is similar to the "bends1,1 but has important differences.
The one principal and significant difference is that fish held
in supersaturated surface waters readily develop symptoms of
gas bubble disease and lethal emboli with no change in pressure
or temperature. Further research may determine that an embolism
in fish may be caused by aziy one or a combination of the factors
discussed above.
-------
26
Chemistry and Physics
Gag super saturation in water can occur from a temperature
increase such as in the thermoclinc of a lake (Harvey} 1967)
or from turbulent mixing of air with water (Harvey & Smithj
1961). A turbulent stream can have a supersaturation by air
injection, and this is observed to occur in the open ocean,
presumably the result of wave action (Craig & Weiss, 1971).
Ground water can be very high in dissolved nitrogen and may
cause problems when used for fish culture (Marsh & Gorham, 1904).
The problem of gas supersaturation uf primary concern on the
Columbia and Snake Rivers is caused by the entrainment of air in
water as it plunges over the spillways of dams. The air bubbles
are carried to considerate depth in the stilling basin (tail-
race) of a dam, and the increased pressure forces some of the
gasses to go into solution. In accordance with Henry's Law,
gasses enter solution j.n relation to the products of their
partial pressures and solubilities. In air-equilibrated water,
oxygon has about one-half the concentration of nitrogen because
oxygen is twice as soluble as nitrogen but in air has only one-
quarter c-f its partial pressure.
Air bubbles carried to depth in water become smaller as the
water pressure increases and Uheir gasses enter solution. As
the bubble becomes smaller, its internal pressure becomes greater,
thereby forcing the gasses to go completely into solution. The
nitrogen would enter solution four times as plentifully as oxygen
instead of twice as readily as under surface or atmospheric equilib-
rium; and nitrogen saturation, expressed as the percentage of gas
saturation would tend to exceed that of oxygen. Temperature can
be an important factor in gas supersaturation. For example, the
solubility of nitrogen at 0 C is 55 percent greater than at
20 C (Weiss, 1970B). Thus, warming of water, without corres-
ponding rc-equilibration with the atmosphere, as on the surface
of a reservoir, can cause significant supersaturation.
Methods of Analysis
There are fout; methods used to study the solubility of gasses
in liquids. These methods are briefly described as follows:
1. Mnnomctric-Volumetric methodsa The most extensively used
methods measure an amount of gas as it is absorbed into
-------
27
n ch'}vt".!;ccl solvent (Oslw.'ild method)u At present, this
In nl-iii tlic imdcI ni-cuiTiU' approach to liio.'it >',ns aolu-
MfJty worka AnoLiifi: turl.hoil cxLracLfi nnd wnburca
(•as fiom a saturated solvent (Van Slykc method)„
2. Mass spectrowetrie. Gas is stripped from sample and
analyzed by mass spectrometry.
3. Gas chropia trographic« Gasses from a liquid sample are
extracted, separated in a chromatographic column and
detected, usually by differences in thermal conductivity.
4. Chemical„ Not used for nitrogen; the Winkler method
is used for oxygen determinations.
The Van Slyke method of analysis has recently been the
method most extensively used in biological investigations for
determining nitrogen, plus traces of inert gas; however, it
takes 30 minutes to analyze a sample and requires bulky and
delicate equipment that is difficult to niovcu It is used by the
NMPS as a sLandard for the gas chromatogrnph which requires
seven minutes for each sample and therefore more useful for
large numbers of samples, as in the current investigations.
The results are detected and recorded automatically. The need
for equipment that is more stable and more portable has
prompted the development of a prototype portable gas monitor by
the EPA„ It is a modified gas chromatograph which requires four
minutes to run a sample, can be run from a portable gasoline-
powered generator, and has a number of advantages for field
use. A few technical problems prevent its being completely
operational at this time.
Problems of Analysis
Due to the lack of portable equipment for analysis, it has
been necessary to collect field water samples and transport
them to a laboratory for gas analysis. It has been standard
procedure to ice the water sample immediately after collection
to increase the solubility and thus minimize the ex-solution
of supersaturated gasses that may occur between collection and
analysis. This error is believed to be small, however, -some
samples collected by EPA have been observed to have formed gas
bubbles during iced storage in the laboratory prior to analysis.
This gas accumulation may either be the result of suspended
-------
28
air bubl conunji out of suspension, or of supersaturated con-
cent rn M on s in excess of solubility at ice temperatures., It
is in Lores Lin;^ to note LlMt the water samples collected from
3rownlct.' Reservoir did not contain the free gas bubble in the
sample botLLe, while all samples collected downstream from Oxbow
and Hells Canyon did contain the gas bubble.
One theory for the high dissolved gas levels found in the
Columbia River is the presence of microbubbles in the water that
are high in nitrogen concentration and virtually non-equilibrat-
ing (Spink, 1971)„ Work on this phenomenon remains incomplete.
The solubility of nitrogen, based on measurements made
during die past ten years, is known to well within 0.5 percent
(Weiss, 1970), However, most saturation values reported for
the Columbia River studies have been calculated from the solu-
bility data of Fox (1909), which may be as much as 2 percent
in error. As an additional complication, many workers have
calculated the solubility of air nitrogen from the solubility
coefficients using a dry rather than moist atmosphere. Satura-
tion values calculated in this way may be up to 2 to 3 percent
too low, depending on the temperature. Measurements of nitrogen
concentrations by the Van Slyke method require correction for
argon and other inert gasses, which are not distinguished from
nitrogen by this method,, In many cases this has not been done
or has been done incorrectly, resulting in nitrogen concentration
values up to 2 percent too high. In many cases, workers have
reported concentrations in mg/liter rather than in ml (STP) /
liter—a unit which is more desirable since it is closer to the
actual units yielded by the analytical procedures being used.
All of these possible differences make comparison of gas
concentrations from the literature and from current investigations
difficult, and often impossible because investigators have not
reported exactly what tliey have measured. What appear to be
anomalies in the data from any monitoring program may be the
result of any combination of the above factors. Data from the
NMFS program are extensive and intercomparable. In spite of
some attempts to standardize methods, the analysts should clearly
report their methods and actual measured concentrations so results
can be compared with other work. It is apparent that there has
been a lack of communication between the biologists and the
oceanographers working on similar problems but for different
purposes. A joint work conference between these two groups is
necessary to develop standard methods for water sampling and gas
analysis.
-------
29
An nppro.icli t.n tlio supi'Vsnturat i on problem that would have
much moiv nu'T l L from n phys Io] opical standpoint would hu to
measure Cfio tot.il pjrl/nZ press tire dissolved gasses in the
water. It is well pstnblishod tliat all of the dissolved gasses
are involved in the gas bubble disease phenomenon, and measure-
ment of nitrogen concentration alone is not necessarily repre-
sentative of the effect on fish.
A prototype device to measure total gas partial pressure
in situ has been developed (Weiss, personal communication),
however, until such a device is generally available, it will
remain necessary to calculate the sum of partial pressures of
nitrogen, oxygen, and argon from their rsasured concentrations
and solubilities.
-------
IV - BIOLOGICAL SOLUTIONS
The crjtical nature of tlie nitrogen supersaturation problem
on the Columbia River 1ms prompted drastic actions to somewhat
relieve the demonstrated detrimental effects on the upstream
and downstream migrating salmonids during the 1971 season. The
mortality of juvenile migrants from the Snake River has been
amply demonstrated and this fact alone dictates immediate action
to salvage a portion of the downstream migration from the Snake
River. An emergency meeting of the Northwest fishery agencies
to consider the nitrogen problems was held in January, 1969 and
resulted in foi-maLion of an ad hoc committee on nitrogen. A
number of actions have been taken since that time by the fishery
agencies and the Corps of Engineers.
Collection of Downstream Migrants
Experiments on the Snake River by the NMFS have shown that
it is feasible to trap a portion of the downstream migrant chinook
salmon and transport them by truck to the estuary for release,
thereby greatly increasing their survival and not seriously
imparing their homing ability as adults. The studies with marked
juvenile spring and summer-run chinook in 1968 - 70 show a defin-
itely increased survival from 1.5:1 before spill to 4.0:1 after
spill due to transportation of the migrants from Ice Harbor to
Bonneville (Ebel, 1970). A plan for experimental installation
of traveling screens in one generating unit and the trapping plan
at Little Goose Dam has been implemented for 1971 funded by the
Corps of Engineers. The success of the program has been hampered
by mechanical breakdowns. Full scale installation of screens
at Little Goose will be based on the success of the 1971 program.
Adjustment of Hatchery Releases
The usual time of a release for most hatchery fish in the
Lower Columbia River has been May and June, the period of maximum
flow and maximum nitrogen levels. The Oregon Fish Commission
trucked a large percentage of their juvenile fish in 1969 from
hatcheries at or above Bonneville Dam and released them near
Astoria in order to avoid this situation. In 1970, an attempt
was made by all agencies to release at least some of the fish at
the hatcheries before nitrogen levels become too high. However,
-------
31
an unexpected and early increase in the river £low made it
necessary Lo release most of the hatchery production of all
fishery agencies into water of relatively high levels of nitro-
gen. In order to avoid this proble;n in 1971 and make the best
possible conditions, for the release of hatchery fingerlings,
a closely coordinated plan between the Corps of Engineers and
other power producers and the fishery agencies was developed.
This was accomplished through the efforts of an interagency
task force that met bi-monthly during the spring of 1971.
Careful advance planning on the part of the fishery agencies,
the Corps of Engineers and Bonneville Power Administration
was necessary in order to implement this plan. A mass release
of 40 million fall chinook fingerlings by the several fishery
agencies was made on April 26 and 27, during a period that
the flow at Bonneville Dam was held to 180,000 to 200,000
cubic feet per second. This flow minimized the spill at
Bonneville Dam and eliminated spill at all other mainstem dams
on the Columbia. Spill at the three lower Snake River dams
was reduced during this period. Nitrogen conditions below
Bonneville Dam were minimized during this five-day period of
reduced flows and indications are that the fish moved out as
expected.
A release of steelhead fingerlings from Dworshak hatchery
on the Clearwater River in early April failed to produce the
desired results probably due to cold water and small size
of the fish which delayed migration.
The wild fish emigrating from the upper Columbia and the
Snake River will receive only limited benefits from the con-
trolled release plan as they characteristically move on the
highest flows which is also the period of the highest nitrogen
concentrations.
Future of Up-River Fish Runs
There is considerable pessimism over the current nitrogen
problem and its effects on migrating salmon into the Upper
Columbia and Snake Rivers. If the studies of the NMFS on
the Snake River accurately portray the fate of wild migrants
from t?iat area and current efforts Co alleviate the problem
are only partially successful, the Snake and upper Columbia
runs could be greatly reduced within a three-year period. The
-------
32
runs of £ii 11 chinook salmon above Priest Rapids Dam are already
in jeopardy because of Lhe failure of artificial propagation
facilities (spawning channels) in that reach of the river
(Meekin, 19/1). Dams and reservoirs have converted the Columbia
River from a flowing stream into a series of lakes (with ex-
ception of Lhe Hanford reach) which retard the out-migration
of juvenile salmonids to the ocean and thereby subject them
to increased stresses including temperature, diseases, preda-
tion, and high nitrogen levels. The fate of juvenile migrants
from the sizeable fall chinook population spawning in the
Hanford Reservation below Priest Rapids is unknown.
Mortalities to juvenile salmon and bteelhead passing
through the turbines of dams is well known. Better control of
discharge by increased upstream storage and passage of more
water through the turbines will aggravate this already serious
problem. The only hope to reduce mortalities front this source
is implementation of the fish screening and trapping program
that has been initiated on the Snake River.
If n major percentage of the juvenile migrants arriving
at Little Goose Dam on the Snake River can be successfully
trapped and transported to the lower Columbia River, they
can be protected both from Lhc high nitrogen and high mortality
from passage through turbines. The high natural potential for
producing these fish remains in the Salmon River where no dam
development has taken place and in several other smaller tribu-
taries to the Snake River.
In summary, as long as unacceptable nitrogen levels (105
percent) exist and high turbine mortalities occur, it seems
the only reasonable way to salvage a portion of the Snake
River salmon and steelhead runs is by trapping the juvenile
migrants and hauling them around the area of high nitrogen
from Little Goose Dam to below Bonneville. An accelerated
schedule for installation of the trapping facilities will be
necessary to accomplish this. The future of runs on the main-
stem Columbia above the Snake River is less certain. Collection
of migrants at McNary Dam would help to maintain those runs
as well as collect the residual from the Snake River. Imple-
mentation of all controls for nitrogen will help relieve
exposure of migrants from tributaries below McNary as well
as give considerable protection to adult upstream migrants.
-------
V - ENGINEERING SOLUTIONS
Introduction
Purport • _nt u 1_ S_c ojh;
The purpose of Lhis chapter is to present and discuss engi-
neering solutions Lo the nitrogen problem in the Columbia and
Snake Rivers. Since excessive nitrogen supersaturation in the
river pppitars to be directly related to the quantity of flow
(usually flood Clows) passing over the spillways of dams and
plunging into deep stilling basins (sec Figures 6 and 7), this
cK'.M.or will deal primarily with the causes and effects of
these spiles and vill discuss proposed methods of i-educing them.
The magniLudc and duration of spills expected in the existing
system of dams and reservoirs in the Columbia and Snake Rivers
will be examined under varying hydrologic and operational condi-
tions. Conditions expected to develop in the future will also
be analyzed.
Concr.iv.iMy there will lie solutions to the nitrogen prob-
lem wlii rh would lu> pnvsically -possible but, fccc-rmso of operational
or le|;al resttii tiuns, would bo unavailable lor actual imple-
ments tier. Tiicto constraints to solving, the problem will be
identified and the problems in removing these restvictions will
be discussed.
As demands for power and water continue to grow in this
region, power generating facilities will have to be expanded
and reservoir regulation operations will have to be modified.
The effects of this expansion and modification on the nitrogen
problem will be discussed.
Finally, recommendations for engineering solutions will be
made, and needs for further study of some problems and potential
solutions will be identified.
Methodology
Several basic assumptions are necessary to an analysis of
oiij\iiiocriny aspects ol" the niLrogcn problem. As stated in
Anonymous, 1071C, these are as follows:
-------
Figure Dissolved Nitrogen Relationship—Lower Snake River (Little Goose Project)
-------
iOO
BOHAJ£VtLL£ (?o)
8o*j*/cvui£ ((>7)
jL
8CwJCU/il£ &/)*¦} ft>j0 8£L0UJ~'
0 80U*J eviLie (fiAtZ^JOU n)
(70) A O^cuate (MA? -Juty/oj
0 doeacTr (ha/ -Jvty 70)
R.AKHeg. juiy 70J
so
/oo
j so
zoo
zso
300
3SO
<0O
PoWFy/iir iU,'/"/ D.'££t;AZ.ZC (_ C C /-i,)
Sf>/Lt u>/>y o/sertJ/Vz.
-------
36
"A. The spillway-stilling basin structure, the release
area for river flow surplus to that required by the
turbine-driven generators of the powerhouse, is the
location where the nitrogen supersaturation originates,
B„ Flow released through the power plant does not appear
to increase the saturation level of the incoming flow
to (.he project.
C„ The output nitrogen supersaturation level downstream
from most spillways generally increases as the flow
increases -- possibly up to a certain level and then
remains about the same.
D. There is some indication that the deeper stilling
basins are more "efficient" in supersaturating the
project outflow."
To provide a framework for an engineering study of the nitro-
gen problem, a "base condition is first described and analyzed.
This "base condition" consists of the existing dams and reservoirs
and those under construction (see map, Figure 1), existing power
facilities and those expected in the future, and normal operating
procedures which would be followed if no particular consideration
were given to solution of the nitrogen problem. Using computer
studies of reservoir regulation and power generating operations
prepared by the Corps of Engineers and the Bonneville Power
Administration, the characteristics of spills expected under
the "base condition" and during the normal flood season of low,
average, and high water years were analyzed for various locations
in the Columbia and Snake Rivers.
The computer studies used in the analysis are based on
historical hydrologic data, synthesized from actual flow records
of past years and adapted to account for the present and future
regulation of the rivers by the existing and future system of
dams and reservoirs. Based on the hypothetical unregulated vol-
ume of runoff at The Dalles, Oregon, occurring during the flood
season (April through August), the year 1956 was selected as a
year with relatively high flows, the year 1952 was selected
as an average water year and 1944 was selected as a relatively
low flow year. Flows comparable to the 1956 flood have a
statistical probability of being equalled or exceeded roughly
onco in ten years. Average flows, comparable to those in 1952,
would be equalled or exceeded once in two years„ Low flow condi-
tions (1944) have a probability of being equalled or exceeded
-------
J /
of 95 percent.; or, only one yrtn in twenty would lower flow
be expected. Although the 1956 flood has a statistical proba-
bility of being exceeded fairly often (once every ten years),
it is still a flood of high magnitude, its volume being the
largest recorded at The Dalles since 1900 (see Figure 8).
After a past year was selected as having a given frequency
of recurrence (based on flows at The Dalles), hydrographs of
flows at the other stations in the Columbia and Snake Rivers
were tabulated and graphed for that same year and assumed to
have roughly the L.ame frequency. This introduces a possible
source cf error m the method, because flows at The Dalles
in a certain past year may have a different frequency than the
flows at another station for the same year. In fact, frequency
analyses of historical flows on the lower Snake River revealed
that flows in the year 1932 were actually much more representa-
tive of average conditions than the 1952 flows which were
selected cs average for the Columbia River, For this reason,
1932 flows are presented in this report as average for the lower
Snake rather than 1952. The method of analysis must be assumed
approximate, but is considered adequate for this presentation.
In the second phase of the study an analysis was made of
the effect on the base condition of various proposed methods
of controlling nitrogen. The solutions studied fell into three
general categories. The first were operational types of modifi-
cations to reservoir regulation and power operations. The second
broad category consisted of solutions requiring structural types
of modifications, and the third group consisted of combinations
of operational and structural modifications. All the solutions
studied in this phase were considered available for implementa-
tion, providing capital and operational costs could be funded.
Legal restrictions on the measures could not be discerned.
The third phase of the study consisted of an analysis of
some types of solutions which were considered physically possible,
but involved some significant legal or operational restrictions.
In this phase of the study, it was first assumed that the only
limitations on the solutions would be physical; that is, there
would be no legal or operational considerations involved,, The'
effecLs in reducing spills were then measured. After this was
done, the constraints that had to be removed to implement this
set of solutions xjere examined to determine the problems and
costs involved.
-------
COLUMBIA RIVXR
ftT
THE DALLES. OREGON
1879 - 1255
K - TO - YOLU"; RELATIONSHIP
PRtL THRC'JCH AUIUST VOLUUE
floc3 ccitjcl cfEuni'; rui
coluuu eitlb t:uti
JJ'.T ISU
60 70 80 SO 100 I
X = A?PJ'_ THROUGH A IGUST tinRfftULATF-D RUNOFF VOLUUE A3
iO 120 130 140
THE OALi.ES IN IGOOfXO AC. FX
Figure 8 Unregulated Flow at The Dalles, Oregon
\J Reference: Anonymous, 1968.
-------
39
Because of tunc limitation the analyses described in this
section were based ahnost exclusively on data and computer
studies already on hand and usually prepared for other purpose?.
No attempt. was made to perform additional studies specifically
tailored for nitrop.cn considerations. Consequently, information
had to be drawn from several sources, with some sacrifice in the
continuity of certain presentations.
Analysis of Lhc "Base Condition"
Definition of Base Condition
Problems related to nitrogen supersaturation have been noted
in many streams throughout the Columbia-North Pacific Region.
However, for this study primary emphasis was placed on the main
stem Columbia and lower .Snake Rivers. The stream reaches selected
for detailed study were: the main stem Columbia River from its
mouth to Grand Coulee Dam, at river mile 596.6; and the Snake
River from its confluence with the Columbia River up to Lower
Granite Rim, at river mile 107,5. Although some information was
collected on all the clams in the reaches listed above, due to
the limiLod scope of the study and time limitations involved,
certain dams and reservoirs were selected for detailed presenta-
tions of hydrology and power operations. Problems at other sel-
ected dams in the Columbia-North Pacific Region were also noted,
however, and will be mentioned.
Average daily flows obtained from Corps of Engineers' com-
puter studies were tabulated and graphed for two control points
on the main stem Columbia River — Priest Rapids and John Day
dams -- and for one pjint on the lower Snake River -- Clarkston,
Washington. These graphs were constructed for the conditions
existing as of 1970, and for conditions expected in 1980, and
used with predicted flows through hydroelectric turbines to pre-
dict spillway discharge. Each case was examined for effects
in high, average, and low water years corresponding to historical
flows in the years 1956, 1952, and 1944 respectively,, Some 19
million acre-fceL of active storage in three new reservoirs --
Libby, Mica, and Dworshak — will soon be available for river
regulation, and the predicted 1980 conditions reflect the pro-
posed ubc of this additional storage. Libby Reservoir is presently
in Lhc process of filling, Dworshak is scheduled to begin fill-
ing in 1972, and Mica is to begin filling in 1973. By 1975 all
three should be operational.
-------
40
The projects included in this study and their hydraulic
capacities arc shown on Figure 9. The dams shorn are all
federally owned and operated except for five located in the
Mid-Columbia River reaches0 These five — Wells, Rocky Reach,
Rock Island, Wanapum, and Priest Rapids -- arc owned by several
public utility districts located in the State of Washington.
Although Lhe projects listed were the ones selected for analysis
in this study, the entire system of dams and reservoirs on the
Columbia and Snake Rivers and their tributaries was included
in the computer studies used for predicting the flows at
various points.
Figure 10 presents hydro^raphs of flows during the flood
season (April Lhrougli July) at selected projects in the Columbia-
Snake River system, and expected flows through Lhe turbine-
generator units at each project under the "base condition,,"
Any flows above those required for power generation are assumed
to pass over the spillways.
One set of data (Figure 10a) presents the situation as of
1970; the oLher set (Figure 10b) shows the situation expected
in the year 1980 when additional river regulation is available
from upstream storage in the new Libby, Mica, and Dvorshak
Reservoirs. No other significant storage will be constructed
before 1980. Therefore, the hydrographs of the 1980 situation
represent conditions from about 1975 (when the three new reser-
voirs are operational) to 1980,, On the other hand, in response
to an increasing demand for electric power in the region, the
loads carried by the power facilities at the dams will also
increase. As the system loads increase, more generating units
will be added, more flow will be passed through turbines and
less therefore over the spillways. It.is anticipated, then,
that average spillway flows will continue to decrease up to
1980. Power peaking (discussed later) will be increasing,
however, and this may cause the daily peak spills to increase
through the period.
Given the present hydraulic design of the turbine units
in the Columbia system, it is not practical to allow them to
pass water without generating power. The power units in the
Columbia and Snake systems are designed to operate best when
they are generating about 80 percent of their total capacity.
Running flow through these units with low loads produces
-------
400
tn
*!—
o
O
O
O
CO
LlJ
o
<
CL
<
o
o
_J
<
q:
q
>-
x
300
200
100
0
30
24
COLUMBIA RIVER PROJECTS 2/
LEGEND
-JJ?——NUMBER OF GENERATING UNITS
j {--POSSIBLE FUTURE CAPACITY
I H--CAPACITY BY 1980
I [—EXISTING CAPACITY
40
20
I 6
I 6—
! i o
3 i
iji f
27 I
I 6
I I
*
I 6
ft 7
; !
1L
I 0
1
I 0
I 0 I
,
%
*26
-
%
Ho.
%
m
I 4 31
1/
I 6~
CiCT.Trvn
H
f—
i"l
A
,
a
'<0o '
%
*
1/ Presently under Active Study
_2/ Reference: Anonymous, 1971C
LOWER SNAKE
RIVER PROJECTS
(1983)
6 6 6
i i i
Gn %
vV
<&-
0
%
October 1970
Figure 9_ Hydraulic Capacity of Generating Units
-------
Figure 10a(1) Base Condition, 1970, High Flow Year (1956)
loictl t/onj
-------
Figure 10a(1) [Continued] Base Condition, 1970, High Flow Year (1956)
-------
ii Mf!l-
"i' =!: :I!:!i|I !!:
V-¦ i!i;:
!»!!!|!iis
I ::
ii i.; ! i! i
¦: 1:1!, 1
.. ;'V:-'i i!i:iii11 ¦ |i
fiiJ liji'ii1! 1 * M l|
£ty!|ii!j;a!p>;^ Uljiil
,:,;i !-'i?li'lii iliVijiyljii©!
|i; pum ?
'i;;;! Priest Rapids
I |: 11 '
L/^-- 'KJ\ , i "j'V,' j
I t
!¦:
.•!! !iiij?ni!ii:li!li!!i Mm'
i1 • • i j. • 1 i
11 ii11i!:!
;j \\\\
! |;' 1 i i i 1!
i t .i,!
111;.:;, ;
! |: •! 11 j 11
:Mi! iii !
; ' j j ;! / I
iiipii
• !
• stain
! ' ' ;
mm
llllll 1
/
AT
•
•T
aT «»(~ ;
/r _ jej/ at
Apr,/
/Ujf
Ju/IC
! 1:! '» • |1
J"u/y
iii;'1
• ; • : ' 1 • \ ? 14 i
U
j . iSt
3C Li f "f/e Goose $
Loujer N\ OIKJhien fa/
1:1
L SC
Toia! -f/ouj
passing projccT-
¦ d' ffc/xncc » j=/*//
Mydraaft'u CapQC'^y
of turbines
FJoui -through
-tufbi'ntt
/»»o«VAA^3
:/ t'S *~ / . is ;. . «?/1 ~ / 5" Jo, /
.• sec
¦l: '
fP fT
i •
I::':
K^.
« ISO
j5 f •
» «:.!*!!!
! '• • I! i': 'i
(11!; i; m •
i!;; •:
i! ''! il:;
j ; :s i'-;':
iii'i' .'i-
,1
; '.-isji.
. I H | 1
I i i i I i!
II |l|
i. i:!
illi!
is.
' '
<) 1 | j ;
iiiiiji
i' j ij"'
ijjli 11
:!;|;i!
V:T
¦Tr
i; • i
! ! : 1 j!
iii',:'!
\ ';!' !
^ i • »
!
Roc k Island
' ' : I . • i' • •!::
/ ..!"«•* \ .'J I,! _
AM ii ii l&l L- i!iiilii i1Ii5 MlMi ©iiiiliiltoiiiii
; ¦'' :' i' ¦ •! ': • i ¦ i ¦ •'; >:; j; i; 11' j i' j; ~ ;; ¦: H"[! pT:: ¦; ; •! ¦!,!
!;;ii!|i!|.;l!iii
,i¦!'; i! '- i1 -
! :: ¦! '--'lliiijiili!:!!!:!!
Ar-I
-F
• i i i!' i'
II;! !
1 ;»•:*• I
. ,, . -7--n
':; I: j1: • I;, i;1 • i:11 •' i'
'!!•; • . :;; :'.11 :.1: . .
• ' i.i ':!:! 1
J*
JufK
J*r . /}
¦Tu/j
—1
^ ::: '
Figure 10«(2) Base Condition, 1970, Av*r*g* Flow Year (Columbla-1952, Sivake-1932)
-------
• I . r* ** ' f" ' «"• I I .TrTr* Tt.T'W-v-— ; I,:-
i! ¦ *:: i i • " '!'•!'! •1' 1 I}' i 11 i
•!M it:i1 j¦'¦ i' ':¦!:";ii! '!!!¦: ill
i • i , •, '. . • • i; • ', •. • ; • |: ; I •!;
- lil[ii|iiiPill|i!l McWar>'
'L: / v; ¦' '¦¦¦ ¦
I "
I • '
i
i •
t • ! .
f
•
i 1
iii: i
|*i . ! .«;«{}••. i= i; :t!|!;j.;i!::' ,
' ; '¦>' :r-',",:p" ::'-i!]•:;!": j Jofi n Day
;I!:!il!• iI!ii!:!;j• ii
¦: •' *"n :: : ii '
n;uil'tS
1 | _!¦I :: ' j j
ill
\ ! -I'l
Ill i„$L .11« ' H ; ] I" : . i . I •
i ;-.i •. ip mi:.-.; :¦ : ;•
IS . VI I
IS tn
Ap*il
H *y
Tu ty
if::;:
j I j!': '
ill..
Lil!'<
fl|!:
500
HS -
j i! ^
' ' o 30i
r.' ®
iiiiC
I . •
lit-
!|•
ill:-;
Figure 10«(2) (Continued] Base Condition, 1970t Average Plow Year
(Columbia-1952, Snake-1932)
Total -f/oto
passing orojttJ-
My
-------
Total -f/ouj
-------
onne vi
Fisure 10a(31 (Continued] Base Condition. 1970. Low Flow Year (1944)
-------
Figure 10b(1) Base Condition, 1980, High Flow Year (1956)
iiii
Total f/ouj
pats/nj proj
Li i+le &
oose
[I ;
II ...U ' rs ! '
li-iiitifil:'! ul ;! W¦ \ i
; ' • " " • •»~. /. i ' \
III " f\ I / j.lHiV
l1! |i /!•:'!
/Aj i i' i \ ¦ i • 11!!!!! ¦ 111': • J J i i ¦ i i i | ;
b ' : WilM "1 »ili !
11 i& II l!I |!1 illlj IlihH
i[i|sji!;|!lp!;i!
• : |
, i 1 :i. , •
iiiiliM'
•: i '•','! 1',! i :' i. ' i ^ —•
| j!!!!! ^! i i i i • •;;• s1 •! ¦!:;1 :: ^
I a . *j/ // ";| jtj/ if i / ¦ /£
I'
4*"'/ . .' • //c\ ¦ .,!
iilh:; Ice Harbor
ifi :
I H !;;! ;SlijPii
'iHllliilSilliiilil i|!i| iipijl Ipiijili^HSi-O;:
liililililjij¦!! Ipil iilijt
I!;!'!:'!!::1!:;:-!'!!!::!!;!'!!!!'!! |l|!ii|[rr•iUijjl;" !'! : !-:
1 br: r-
j . .
I • ,
:i: l:
11 •; • ¦ •' j ~, ;1: -.
>t 4-y :
Afrtt ! . J~uj*€
¦*T
-------
To^af •fcfottj
post/nj pcQj^vt
/ujc/rau/zL Capacity
of "turbines
FIchaJ 4hrcujJ\
¦&ur6i'nes
/»»o*tVAA^3
Lcgcnc/
' 3/>'//
/"?K:
Figure 10b(l) [Continued] Base Condition, 1980, High Flow Year
(1956)
-------
-------
Figure 10b (2) [Continued] Base Condition, 1980, Average Flow Year
(Columbia-1952 Snake-1912*
7o/o/ -fi/ouj
pass/nj proj« /
fJycJrauft't. Capacity
of turbines ~ — —
Floui ihroujjh
tuf hints
¦ c/i V/ercocC * S^it/
Lcjcnc/
o rtnevi
•J V LY
-------
Figure 10b(3) Base Condition, 1980, Low Flow Year (1944)
To /a/ -£/ouj
-------
Figure 10b(3) [Continued] Base Condition, 1980, low Flow Year (1944)
7o/a/ "ffouj
pa&s/nj proj*<.S-
£/ouj ihrovjh
tuf hints
Legcna/
-------
54
localized high negative pressures and leads to the phenomenon
known as "cavitation." The high negative pressures cause tiny
bubbles of water vapor to form and when they collapse (or
"implode") in regions of higher pressure, the force of the
implosion creates stresses on turbine blades and other sur-
faces in the water passages. Operation at low loads has also
been shown to cause more damage to fish which are passed
through the turbines during their downstream migration. For
these reasons it is assumed in the "base condition" for 1970
and 1980 that the flows passing through the turbines at a
particular hydro project are limited by the amount of energy
load to be served by that project. The total hydraulic capacity
of the turbines, of course, defines the maximum amount of flow
which could be passed under any conditions. As shown on the
graphs in Figure 10, the flows which pass through turbines at
a project are normally less than the total hydraulic capacity.
This is done to maintain some generating reserves to provide for
unscheduled or forced outages, scheduled outages for maintenance
and repairs, and contingencies for unanticipated load growth.
The amount of reserve varies among projects.
It is the total energy load then, not the total hydraulic
capacity, that presently determines how much water can be
passed through the turbines. It follows, that addition of
generating units at the various hydro projects prior to corres-
ponding growth in regional power demand will not help solve
the nitrogen problem. In fact, premature installation of the
units may eliminate flexibility in the use of empty turbine
bays to bypass flow (discussed later) and thus may actually
aggravate the problem. Figure 11 shows that, unfortunately
from a nitrogen standpoint, the period of lowest power demand
annually coincides with the period of highest stream flows.
The curves representing median water and critical low water
conditions terminate where they meet the total energy load
curve. Any flows above those points could not, on the average,
generate power and would pass over spillways. In a later
section a method of designing powerhouse facilities to pass
flows while generating very small amounts of power is described.
If this measure could be implemented, existing generating
units plus new ones could contribute significantly to reduction
of spills and nitrogen levels.
The flow hydrographs shown on Figure 10 should be viewed
as only an example of the flow patterns which could be associated
with floods of the three magnitudes considered. The hydrographs
of floods in future years could have a completely different
shape than the ones presented in this report.
-------
55
/S
| I I I M I
1 — AVARABIE GEK31ATNG CAPACITY
41
/xS£7VES J
TOTAL PEAK LOAD
7
/t
5—'' iT 1
/
~~T
• ^ i
TOTAL LOAD
\
RW FZAK LOAD "Sx.
^ r *
J.
5?1
"Nb
\
V,
y\
FIRM LOAD
QrtF£/J\ LOAii
—i_>
J.
\
\
\
\
\
UMREGOLATfn FLOW
POWRCUTPUT
MEDIAN WATER
\
ll>
/
/
S O N D J
Mon
"7*"
y cm
-A
/
r^r-
/
£
/
4
/
/
/
1CA1 WATS
M A M J
3
Figure H Annual Loads and Resources, Pacific Northwest—1
1/
1/ Reference: Anonymous, L971C
2/ Converted to equivalent power production.
-------
56
Description of Conditions in 1970
The series of graphs shown on Figure 10a for the 1970 level
of development reveal that even in a year of average flows the
magnitudes and durations of flows passing over spillways of
dams in the Columbia and Snake Rivers would be great enough to
produce nitrogen levels considered hazardous to fish. For
example, in an average year at McNary Dam (1952), spills of
175,000 cfs could last for about two and a half months. Using
Figure 7 as a guide, these spills could produce nitrogen super-
saturation levels of about 135 percent. Evidence presented in
the "Biological Evidence" section of this report indicates that
nitrogen levels above 105 percent produce symptoms of gas bubble
disease in fish and that levels above 120 percent are lethal.
Spills of 30,000 cfs or more at the lower Snake River Projects --
Little Goose, Lower Monumental, and Ice Harbor -- could last
for about two months in an average year (1932). The data in
Figure 6 indicates this level of spill would produce nitrogen
supersaturation levels of 120 percent or more. In the low
flow year (one which would be exceeded 95 percent of the time),
spills and nitrogen levels would be low enough to be considered
safe for fish migrating up and down the rivers. Although it
is not certain at what specific river flow levels nitrogen
problems occur, it is apparent that the threshold level is
somewhere between low and average flows.
The levels of nitrogen supersaturation produced by given
levels of spillway discharge are based on very limited data and
in the discussion above and in following paragraphs should
be considered as only rough estimates of the magnitude of the
problem produced in each case. The actual level of nitrogen
supersaturation below a particular dam at a particular time
depends on many factors, the most significant being the amount
of spilling and nitrogen entrainment in the river system upstream
of the given dam as well as the spill at the given dam.
The water moving downstream through the series of pools
behind the dams in the Columbia and lower Snake Rivers circu-
lates insufficiently to rid itself of the gas entrained at an
upstream spillway. On the contrary, as the water mass moves
downstream, picking up more gas at each successive spillway,
the level of supersaturation often tends to increase. It is
apparent that detailed studies will be required to develop
reliable relationships between spill conditions throughout the
river system and nitrogen levels at particular projects. Along
-------
57
this line, the Corps of Engineers has had developed under
contract a mathematical model for predicting nitrogen levels
produced by various spillway-stilling basin configurations
under given combinations of spillway and powerhouse operations.
Predicted Conditions in 1980
Figure 10b shows the hydrographs and flows through turbine-
generator units expected under normal operating conditions (no
nitrogen controls) in the year 1980, This set of graphs shows
that, if present planning forecasts and construction schedules
materialize, by 1980 there will be significant reductions in
spills throughout the Columbia and Snake River systems,
However, nitrogen supersaturation will continue to be
a problem.
In the operation studies of the 1980 level of development
it was assumed that only three generators would be installed
at Little Goose. However, recent installation schedules call
for all six units to be installed in that project prior to 1980,
and the following comments are based on this revision.
In both high and average flow years, spills at many projects
in the main stem Columbia River will be at levels high enough
to produce lethal concentrations of dissolved nitrogen. For
example, at McNary spills of 175,000 cfs or more would last
for two months iti the high year and one month in the average
year. These spills would produce nitrogen supersaturation
levels of about 135 percent, depending on fotebay conditions.
On the lower Snake River, under the expected 1980 condi-
tions, Figure 10b shows that total river flows above 100,000
cfs could last for about four and a half months in the high flow
year and about one half month in the average year. At the
new lower Granite Project these flows would require spills
of over 30,000 cfs. The levels of nitrogen supersaturation
which this spill will produce at Lower Granite is unknown at
this time. Studies are currently being conducted by the Corps
of Engineers to install additional bypass capacity and to devise
a spillway configuration for that project which will not entrain
excessive amounts of gas. The success of this effort is essen-
tial to the solution of nitrogen problems in the lower Snake
River because the effects of conditions at Lower Granite will
-------
58
carry downstream to the other Snake projects. For example, at
a total river flow of 160,000 cfs, Little Goose Project could
pass a maximum of 130,000 cfs through turbines and would have
to spill 30,000 cfs. If Lower Granite were not entraining
nitrogen, the water arriving at Little Goose might be close
to 100 percent of saturation. The 130,000 cfs passing through
the turbines at Little Goose would also be at 100 percent
saturation, the 30,000 cfs spill would be about 120 percent
supersaturated, giving a mixture some distance below Little
Goose which would be only 104 percent supersaturated. If, on
the other hand, spills at Lower Granite produce a mixture
which is 120 percent supersaturated, the mixture below Little
Goose would be supersaturated by 120 percent or more. The
condition would then worsen down through Lower Monumental and
Ice Harbor since little equilibration would occur in the slow
moving pools between the projects.
The hydrographs in Figure 10b show that in a high flow
year on the lower Snake, total regulated river flows above
200,000 cfs could last for about one month. Rough calcula-
tions show that, if the flow below the new Lower Granite
Project were not supersaturated with nitrogen and all projects
were passing maximum flows through turbines, the resulting
spillway discharges could produce nitrogen supersaturation
levels as high as 135 percent below the most downstream project
(Ice Harbor) during that period. In an average year on the
Snake, total flows of 160,000 cfs or more would last less than
a week and would produce under conditions described above, a
lethal level of nitrogen supersaturation (120 percent or more)
in the mixed flow below Ice Harbor, Flows of 130,000 cfs or
more would last for about one and a half weeks in an average
flow year. Under the above conditions, a flow of 130,000 cfs
would require spilling only at Ice Harbor, and nitrogen con-
centrations in the mixed flows would be below hazardous levels
(105 percent of saturation). Directly below the Ice Harbor
spillway, however, that river flow would produce a supersatura-
tion level of 110 percent or more. This condition would be
less hazardous for the downstream migrants, most of which would
tend to flush out of the supersaturated zone before gas bubble
disease could occur. Upstream migrating adults, on the other
hand, are attracted to the spillway flows and can remain trapped
in the zone of supersaturation long enough to contract the
disease.
If Lower Granite also increases nitrogen levels, conditions
downstream will be correspondingly worse than stated above. Again,
-------
59
iti the low flow year, there will probably be no spills and no
hltrogen problems in either the Columbia or Snake Rivers.
The foregoing analysis of 1970 and 1980 conditions pre-
sents the situation as it exists today and as it would develop
If no attention were given to the nitrogen problem. As the
Bystem of dams, reservoirs, and power generating facilities
expands through the years, the volumes of spill at the various
projects will be reduced. However, without additional con-
trol measures the reductions in volume of spill, even in an
average year, will not be great enough to reduce nitrogen
supersaturation to levels which would be considered safe for
fish.
Proposed Methods of Controlling Nitrogen. Within Normal Constraints
Operational
As noted before, one broad category of solutions to nitrogen
supersaturation problems are designated "operational." These
types of solutions are those which can be implemented through
adjustment of reservoir filling and release schedules, through
transfer of power loads among the various projects, or any other
methods which do not require structural modifications to dams
or power units. Several solutions of this type were partially
implemented by the Corps of Engineers, Bonneville Power Admin-
istration, and the state and federal fisheries agencies in an
effort to reduce losses to fish during the exceptionally high
flows experienced this year. (Based on current forecasts the
flood occurring this year would be equalled or exceeded once
in seven to ten years.)
The complexity of the flow regulations and power generation
network in the Columbia-Snake River system, and the rapid rate
at which system components change from year to year make it very
difficult to select a particular year or set of system condi-
tions for an analysis of the effectiveness of the various
methods of controlling spills and nitrogen. However, since
storage in three major new reservoirs Libby, Mica, and Dwor-
shak -- is expected to be operational by about 1975 and since
no significant additional storage will be constructed between
then and 1980, the effectiveness of the various engineering
measures will be related to the expected river regulation in
the 1975-1980 period.
-------
60
The first operational type solution studied was a modifica-
tion in the use of the five million acre-feet of active storage
behind Grand Coulee reservoir. Preliminary computer operation
studies have been made by the Corps of Engineers to determine
the effects of this type of manipulation on downstream river
flows. Figure 12 shows the results of this analysis. The hydro-
graph of the 1933 flood at The Dalles project was used in this
analysiso The 1933 flood was slightly larger than the 1952
flood; it, therefore, has a frequency of recurrence of somewhere
between one in two years and one in ten years. The horizontal
line at 450,000 cfs designates the flow at bank full stage.
Flows above this stage would begin to flood downstream areas.
The curve labeled "Current Operating Plan" would result
from the regulation of Grand Coulee Dam according to rules
established in the proposed Flood Control Operating Plan for
Columbia River Storage. The curve marked "Minimum Spill Plan"
represents the flows which would result from an operation of
Grand Coulee Dam to provide the greatest control of spills
for the longest period during the spring freshet season. The
criterion for deriving this operation plan is the assumption
that a large spill of short duration is less damaging to the
total fish resource than prolonged spills at lower levels.
Therefore, this operation plan is designed to control river
flows to a maximum of 250,000 cfs for as long as possible, then,
after the storage throughout the system is exhausted, to let
the excess flows go in one short high peak. An approximation
of the magnitude of spillway discharges under these two oper-
ating plans can be made by comparing the hydrographs in Figure
12 with the hydraulic capacities of the hydroelectric projects
shown on Figure 9, and assuming the projects could be operated
at their stated capacities. It may also be assumed that the
flows at the other downstream projects — McNary, John Day,
and Bonneville — are roughly the same as shown for The Dalles.
At McNary, for example, the "Minimum Spill" plan would result
in a spillway discharge in excess of 20,000 cfs for a period
of nearly two months, producing (depending on upriver condi-
tions) nitrogen levels above 120 percent of saturation during
that period. Under the "Current Operating Plan" the same
level of spill and nitrogen supersaturation would last about
two weeks longer. At John Day and The Dalles, the hydraulic
capacity of the turbines will be increased to about 350,000
cfs by 1975. Both operating plans would therefore result
in spilling at those projects for about fifteen days. The
"Current Plan" however would allow spills to be held to 10,000
cfs, whereas the "Minimum Spill" plan would allow a peak spill-
way discharge of 170,000 cfs. At Bonneville, until the six
-------
Sank Full Flow.
nrr,urn
JUNl£
Figure ^ Hydrographs of Flow at The Dalles Resulting from Modified Operation of Grand Coulee Dam
(1933 Flood) y
_l/ Prepared by Corps of Engineers, North Pacific Division.
-------
62
additional generating units are installed (now scheduled for
1978-1980), spillway discharges in excess of 40,000 cfs would
last for three months under either plan. These discharges at
Bonneville would produce nitrogen supersaturation levels above
120 percent during those months. After the additional gener-
ating units are installed at Bonneville the period of these
critical spills would be reduced to one month under the "Current
Plan" and to fifteen days under the "Minimum Spill Plan."
The benefit of the "Minimum Spill Plan" is most obvious at
Bonneville and McNary Projects, where the period of critical
spills can be reduced. This would reduce the chances for
migrating fish to be exposed to lethal nitrogen levels and
would allow hatchery releases of fingerlings to be better timed
to avoid the period of high spills. At Bonneville, however,
the benefit will not be realized until the second powerhouse
is completed and the additional turbines are operable.
There would be some consequences to the "Minimum Spill"
type of operation, however, due to the increased potential
for downstream flooding. The operation may cause flood con-
trol storage in upstream projects to be completely filled
early in the freshet season before the danger of high stream
flows has diminished. The capacity to control later peak
flows is this exhausted. The potential for serious downstream
flooding is accordingly increased. As shown on Figure 12 the
"Minimum Spill" operation results in peak flood flows which
are some 70,000 cfs above bank full stage (for the hydrograph
of the 1933 event).
Thus far all of the computer operation studies used in the
analyses have assumed full use of the storage available in
Canada through provisions of the Columbia River Treaty between
Canada and the United States. In a later section this upstream
storage availability will be discussed in more detail.
Another operational measure which could help to reduce
spillway flows, and thereby reduce nitrogen levels, is a
modified operation of navigation locks at certain projects.
The raising and lowering of the water level inside the lock
is accomplished by opening and closing sets of valves on the
upstream and downstream end of a large conduit located beneath
the lock. There is a possibility that during conditions of
high flow the valves at both ends of this conduit could be
fully opened to allow up to 15,000 cfs to be passed through the
locks. Spillway releases would be reduced accordingly. Tests
-------
63
arc being planned now to determine whether this operation Is
structurally safe and whether the flow released In such a
manner is actually low in nitrogen content. Problems which
could arise with this measure are cavitation in the bypass
facilities, surges caused by closing the valves from a fully
open position, and possible nitrogen entrainment in the
bypassed water through air vents in the valves. A test of
this bypass method this year at Little Goose Project produced
unacceptable surges and vibrations In the emptying valve
shafts. Further tests will be necessary to determine whether
this problem can be eliminated and whether the problem is unique
at Little Goose, however, prospects at this time appear dim.
There may also be conflicts arising during periods when the
locks must be operated for navigation. Locking operations
on the Snake River during the high flow season occur only a
few times per day, whereas on the Lower Columbia they are
more frequent -- 10 to 12 times per day. Each operation re-
quires 45 minutes to 1 hour. The effect of bypassing inter-
mittently must therefore be determined.
Navigation locks have been installed in the Lower Columbia
projects (McNary, John Day, The Dalles, and Bonneville), and
also in the lower Snake projects (Lower Granite, which is under
construction; Little Goose; Lower Monumental and Ice Harbor).
The possibility of bypassing flows at these projects should
therefore be considered. In addition, special studies have
been made by the Corps of Engineers on the feasibility of in-
stalling navigation locks at some of the upstream Columbia
projects (Rocky Reach, Rock Island, Wanapum, and Priest Rapids).
Although a bypass capability of 15,000 cfs is not particularly
large compared to flood flows in an average year; when combined
with other solutions for reducing flows over spillways it
could be a significant partial remedy.
Other modifications to the reservoir regulation procedures
are available for control of excessive spills and some were
implemented during the excessively high runoff period this
year. These other methods involve modification of spilling
schedules during periods of fish passage, varying the pattern
of release through various spillway bays, and establishment
of priorities for spilling at the various projects. These
types of modifications to reservoir regulation can be very
effective, especially with close coordination between the oper-
ating agencies and the fisheries agencies. However, since
-------
64
these modifications lo system regulation tnki* place In response
to critical situations as they arise, it Is difficult to pre-
dict their effects in future years,, Although it is difficult
to quantify the results of this type of coordinated reservoir
regulation, it is apparent now that this coordination will be
an essential element in the success of an integrated solution
of the nitrogen supersaturation problem. This year, for
example, fishery agencies cooperated to schedule the release
of nearly all hatchery-reared fingerling fish in the Columbia
River (40 million out of a total of about 60 million) within
a two-day period. Reservoirs and power-generating facilities
were operated to produce the least possible amount of spill
on those days and for about five days thereafter, to allow
the young fish to move downstream in water which had the lowest
amount of nitrogen possible during this unusually high flow
period. The reduction in spills was accomplished by storing
about 1 1/4 million acre-feet of water in eight reservoirs
throughout the system, and by transferring power loads to
allow the lower Columbia and Snake hydroelectric projects to
generate at or near their full capacity.
Another set of operational modifications for nitrogen
control falls into the broad category of power load transfer.
Since spillway flows at certain dams tend to produce more nitro-
gen than at others, and because some projects are located in
river reaches where large numbers of migrating fish are present,
It is desirable to reduce the spills at these projects as much
as possible and to accordingly allow more spills to pass over
less critical projects.
A vast network of power transmission circuits links gen-
erating plants in the Pacific Northwest to load centers.
Interconnections between generating systems -- public, private,
and federal — provide a high degree of flexibility in distribut-
ing power among the load centers. This flexibility makes it
possible, during periods of high river flows, to transfer more
of the generating load to the hydroelectric plants located at
critical dams. This has the effect of causing the turbines
at these critical projects to pass flows at or near their
hydraulic capacity, with correspondingly less flows over the
spillways. At the less critical projects the power genera-
tion must be reduced and the spillway flows increased. To make
maximum use of the system flexibility the public, private (Idaho
Power Company), and federal power operating agencies and the
Federal Power Commission should be included in any coordinated
effort to solve the nitrogen problem.
-------
65
This type of load transfer was another key feature of the
coordination which was accomplished this year in an effort to
reduce the overall impact of nitrogen supersaturation. The
goal of this operation was to generate as much power as possible
at the downstream projects in both the Columbia and the Snake
Riverso The reason for this was that even though spills at
upstream projects also produce nitrogen, the fish runs from
all the tributaries converge downstream. Therefore, reducing
the hazard downstream has a potential for protecting the
greatest number of fish. It must be noted here that the hydrau-
lic capacity of the turbine-generator units at the various
projects becomes a limiting factor in this load transfer opera-
tion. Also, since many of the dams involved in the transfer
relationships are in public and private ownership, a high
degree of cooperation among the various generating agencies
is required.
Arrangements may also be made to transfer power during
these critical high flow periods to agencies outside the
Columbia-North Pacific Region. The Northwest Power Pool is
interconnected with the Rocky Mountain Power Pool, the Pacific
Southwest, and the New Mexico area. Transfers of this nature
were accomplished this year. One arrangement incLuded transfer
of energy to British Columbia Hydro and Power Authority. Power
generated in the Pacific Northwest Region at critical projects
was delivered to the Canadian utility. This allowed BC Hydro
to store water in their new Wilis ton Lake Project in the Peace
River system in lieu of releasing it for power generation during
the period of high stream flow. This arrangement was only
possible this year because Williston Lake is in the process of
filling. In later years, it is anticipated that this alternative
will not be available for reduction of spills at the Lower
Columbia projects. Another arrangement which was accomplished
this year was the delivery of power produced at critical projects
to West Kootenay Power and Light (The British Columbia utility
whose plants are on the Kootenay River below Nelson, B. el),
in exchange for immediate spill on their system. Unlike the
Peace River arrangement, this transfer had no effect on reducing
the amount of flow passing downstream, but it did allow more
generation of power at the critical projects. Transfers to
West Kootenay Power and Light could be made in future years.
Any load transfer of this nature would be limited by two
factors: (1) The existing and future transmission intertie
line capacity; and, (2) The actual availability of power load
-------
66
in Che regions to which the load is to be transferred. The
effective capacity of the intertie lines to the Pacific-
Southwest Region is normally about 1900 megawatts (two a-c
lines with capacities of 600 megawatts (MW) each plus one
d-c line with a capacity of 700 megawatts). The recent earth-
quake in Southern California rendered the d-c line temporily
inoperable; consequently, the two a-c lines were each increased
to 800 MW, for a total present capacity of 1600 MW. This
situation is expected to continue for about a year and a half
while the d^c intertie facility is under repair. Future plans
include an additional 700 MW d-c line. This could produce a
total future capacity of about 2600 MW. The intertie to West
Kootnay Power and Light has a capability of carrying about
200 MW and the intertie to B.C„ Hydro can carry about 300 MW,
The overall effect of these types of transfers, of course,
is to increase the amount of water which can be passed through
turbine units at critical projects in this region. However,
the fact that the success of the transfer depends upon the
need for the power in the other regions makes it very diffi-
cult at this time to predict the magnitude of the reduction of
spills in the future that could be realized from this type of
transfer. Bonneville Power Administration has indicated that
although this information on predicted conditions is not
available now, computer programs and the basic data required
to determine the effects in future years are available and
could be used to analyze alternative methods of power trans-
fers and power generation for an optimum operation for control
of nitrogen. These studies should begin as soon as possible.
The computer studies on which the turbine flows in
Figure 10 were based, did account for transfers of power within
the United States. This means that, due to present intertie
capacity, the maximum additional load which could be trans-
ferred is about 500 MW to British Columbia. At many hydro-
electric projects in the Columbua system about 7 MW can be
generated by each 1000 cfs passing through the turbines. Trans-
ferring 500 MW of additional load to Canada could therefore
reduce spills by 71,000 cfs. This reduction could be taken at
a single project or split between several, depending on turbine
capacity and spilling priority.
Structural
An approach to finding structural solutions to the nitrogen
supersaturation problem depends on some basic assumptions con-
cerning the relationship between the project design and the
-------
67
levels of nitrogen supersaturation. These assumptions are
stated in iho first pnrngrnph o£ the Methodology section of
this chapter. Figure 13 shows the basic layout of a typical
hydroelectric project.
One very promising structural measure is being tested
for its effectiveness this year. This involves the use of
slotted gates in empty "skeleton" bays of existing powerhouse
units. Many dams in the Columbia and Snake Rivers have a number
of empty turbine-generator bays in their powerhouses to pro-
vide space for additional generators that will be needed in
future years. The actual installation of the generating units
in the empty bays is accomplished on an as-needed basis and
in response to expected power demands in the near future,
A sectional view of a typical skeleton bay is shown in Figure 14.
The intake structure and the tailrace structure are complete,
but the concrete has been omitted in the area where the future
turbine will be installed. Figure 15 shows a similar struc-
ture with the turbine and generator in place.
These skeleton bays are structurally designed to operate
only if the turbine-generator combination is in place and
operating to control flow and absorb energy. Consequently,
until the generating units are installed and operable, the
bays are sealed from any flow by placing concrete stoplogs in
the intake structure and draft tubes.
The Corps of Engineers conducted model tests to determine
whether flows could be passed through these empty bays and thus
help reduce flows over the spillway. The tests indicated that
unacceptable flow conditions and structural stress would result
from this operation unless a temporary means of controlling flow
and dissipating energy were provided. Further tests showed
that a multi-orified (slotted) gate, positioned in the bulkhead
slot in the intake structure, would provide the necessary
control. With the slotted gates in place, the skeleton bays
in projects on the lower Snake River should have roughly the
same hydraulic capacity as they would have with turbine-
generator units operating (about 22,000 cfs per bay).
This year, the Corps of Engineers awarded contracts for the
construction and installation of full size slotted gates in
one of the three empty bays at Little Goose Dam on the lower
Snake River (see Figure 16). Three gates are required per bay.
Results of this prototype test showed that the gates functioned
to pass water with no apparent structural damage to the bulkhead
-------
PECK EL.717
12-50'bays gates sa'Htf.H\ 74c/
r^r
I HIGH \
"709S?
-SEW FlSWAY ENTRANCE
^OENTATED SILL
PLAN OF SPIL1 WAY
SECTION THRU SPILLWAY
_nckval _rrrL_ El ?ct
SPILLWAY 12 t?Vi
\
"HT
I I I
P/H FlSHWAY
'entrance
-S FV» FirHAAT
ENTPASCc
f
V
NORMAL TW EL 612 \
\
p/h fishway
E NTfsANCE
PROJECT PLAN
NOT TO SCALE
HOT TO SCALE
Figure i3 Basic Layout of Structures in a Typical Hydroelectric Project (Rocky Reach)i^
1/ Prepared by Corps of Engineers, North Pacific Division.
-------
fecXi6t GS/-7
b(*M AoiG33
/&/ Cctl£dfWC/MV*J£L ¦—¦>
El £55*7 _ 1
I
rw ei.s&>7
Dk-'s-T7V5£
£vUV/ziiO SioT
Bi 4sz-y
•TjzasMsack
St.t>y
, x'.a
^'dljihv-
Figure _^i. Section Through Skeleton ftaj--LCttle Goo*e Projtct
-------
Figure 16 - Prototype Slotted Gate Being Installed
at Little Goose Project
-------
71
or to the skeleton bay. Water samples collected during the
test by the National Marine Fisheries Service (Department of
Commerce) in the forcbay, skeleton bay, and downstream showed
that water passing through the dam in this manner did not
increase in nitrogen content.
Two other existing dams on the lower Snake River — Lower
Monumental and Little Goose — each presently has three gener-
ating units installed with a total hydraulic capacity of about
66,000 cfs; Ice Harbor has a hydraulic capacity through its three
turbines of about 45,000 cfs. In addition, each of these projects
now has three empty skeleton bays which could, with the modifi-
cations described above, be used to bypass additional water.
If all three of these additional bays were fitted with slotted
gates and allowed to pass water, the flow capacity through each
project would be increased to about 130,000 cfs. Ice Harbor,
with a lower initial capacity would only be able to pass about
105,000 cfs under this arrangement. The advantage of this scheme,
of course, is that each project would be capable of passing
through non-nitrogen entraining facilities quantities of flow
roughly equal to their full hydraulic capacity, without having
to wait until the additional generating units are actually in-
stalled and needed to serve power demands.
Using the hydrographs in Figure 10, the effects of passing
these additional flows through the powerhouse can readily
be predicted. Assuming the lower Snake projects are loaded
to capacity, the magnitude and duration of spilling would be
greatly reduced with correspondingly less entrainment of gas.
However, preliminary analyses indicate that nitrogen levels
will still be high enough during the periods of peak flows to
be hazardous to fish in years with average flows and higher.
After 1975 a regulated river flow of 160,000 cfs or more would
last for about one and one half months in the high flow year,
and for less than a week in the average flow year. Assuming
the total bypass capability were as stated in the preceeding
paragraph, a flow of 160,000 cfs would result in spills of
30,000 cfs at Lower Granite, Little Goose and Lower Monumental
projects and 55,000 cfs at Ice Harbor. If Lower Granite
could be relied upon to add no nitrogen to the system and the
flow arriving at Little Goose had zero supersaturation, the
nitrogen level below Little Goose in the average year would be
104 percent of saturation, but would increase to 120 percent
below Ice Harbor. In the average year regulated flows of
130,000 cfs or more would last about one and a half weeks. A
flow of 130,000 cfs would require spilling only at Ice Harbor,
-------
72
with less than hazardous nitrogen levels in the mixed flow below
the project. However, as stated in the discussion of the 1980
base condition, the supersaturated flow directly below the
spillway would still constitute a hazard for upstream migrat-
ing adult salmonids. In a high flow year, river flows of
200,000 cfs or more could last for about one month. The
resulting nitrogen level below Little Goose would be 113 percent
of saturation and would increase to 135 percent below Ice
Harbor. As stated previously, at supersaturation levels above
105 percent fish begin to show symptoms of gas bubble disease
and at levels of 120 percent or more significant mortality
occurs. Although the hydrographs presented in Figure 10 are
unique for a particular historical year, it is reasonable to
assume that the period of peak flows and high nitrogen levels
described above will occur in any future year at a time when a
large portion of the anadromous fish population is migrating
up and down the river.
Although the "slotted gate" does not solve the nitrogen
problem it does reduce it, and thus increases the probability
that in combination with other measures, nitrogen levels may be
held to acceptable levels. The desirability of this type of
a construction approach cannot be overstated. It offers a
multi-purpose partial solution to the problem, allowing the
maximum in flexibility in use of the empty bays while waiting
for the generating units to be installed and the power genera-
tion needed to meet system demand.
Cost estimates for installation of the slotted gates in
the lower Snake River projects are presented in Table 2.
TABLE 2
Estimated Costs for Installation of Slotted Gates
in Lower Snake River Projects —
Lower
Little Goose Monumental Ice Harbor
Total estimated costs $4,360,000 $4,415,000 $3,400,000
Allocations to date 900,000 0 0
Balance to complete 3,460,000 4,415,000 3,400,000
JL/ Corps of Engineers, North Pacific Division
-------
73
Approximately $1,500,000 of the cost estimate for each
project can be classified as costs for reducing nitrogen satura-
tion. The remaining costs are advance costs related to ultimate
development, involving completion of the powerhouse and installa-
tion of the last three generating units. On this basis, the
total cost, attributable to nitrogen reduction, for modifying
the three existing projects on the lower Snake amounts to
roughly $4.5 million. This does not include additional costs
required at the Lower Granite project, now under construction.
Variations on this scheme have also been proposed, and
some of these variations will be incorporated in the program
to install bypass capabilities at the lower Snake River pro-
jects. One of these variations is called "capped-unit" diver-
sion. With this variation, all the concrete and imbedded
items will be placed in the skeleton bay, including the stay
vanes, wicket gates, and sealed head cover. In this way, the
water passageways can be completed and allowed to begin pass-
ing flow, while at the same time allowing the generator con-
struction and installation to proceed as scheduled. Still,
under this system, a slotted bulkhead will be required to
dissipate the energy in flows passing through the skeleton
bay below. This is a desirable variation at dams at which the
skeleton bays have been partially completed. In dams which
are now under construction another variation is also possible.
This is what could be referred to as "slanted" construction.
With this version, the skeleton bay would actually be designed
and constructed to have sufficient strength and hydraulic char-
acteristics to pass diversion flows through the dam without
the need for energy dissipation in another source such as a
generator unit or a slotted bulkhead. "Capped-unit" diversion
could also be incorporated in this version of the bypass
scheme. The "capped-unit" portion of the construction scheme
is considered very important because it can provide for contin-
uous diversion capability during the freshet season at the same
time that additional generating units are being installed.
The potential for this type of diversion capability at
other projects on the Columbia-Snake river system must be
examined. In the lower Snake system, if funding is received,
full diversion capability will be provided before the 1972
freshet season. At Lower Granite Dam, which is now under con-
struction, present plans call for three turbine-generator
units to be operated when the pool is filled in 1975, with
three skeleton bays which can be used for diversion at that
-------
74
time. On the Columbia River at Bonneville Dam, generating
units are now installed and operating. There is no possibility
of the skeleton bay type of diversion until an additional
powerhouse is constructed„ The second powerhouse at Bonneville
is not due for completion until the years 1978-79, so it is
possible that the capability for diversion may not be available
until those years. (See installation schedule, Table 5, p. 92). At
The Dalles, installation of eight more generating units is
underway and scheduled for completion by May, 1974, It appears
at this time, because of construction scheduling techniques,
there is no chance of bypassing water through the eight addi-
tional bays. A discussion of these construction scheduling
methods is presented later. At John Day, four skeleton bays
are available. These skeleton bays are structurally dissimilar
from the skeleton bays in the Lower Snake projects and may
require additional design modifications before they could be
used as flow diversion facilities. However, diversion through
these four bays should be considered a definite possibility.
McNary Dam has fourteen powerhouse units operable now with
no skeleton bays available for diversion. Priest Rapids now
has ten powerhouse units operating, there are no empty skeleton
bays available now, and it appears from diagrams of the dam
that additional powerhouse units or bypass facilities would
require major construction. Wanapum Project has presently
installed ten powerhouse units with space for six additional
units. The amount of construction necessary for bypass capa-
bility in these six additional spaces is unknown at this time.
Rock Island Project has the lowest hydraulic capacity (about
70,000 cfs) of any of the Columbia River projects at this time,
with ten powerhouse units installed. There are tentative
plans to provide an additional six units at this project,
but judging from sketches of the layout of the dam it would
entail a major construction effort. Bypass capability is then
considered doubtful at this project also. Rocky Reach has
seven powerhouse units installed with four future units.
However, these units are due for completion in 1971. It is
therefore considered unnecessary to provide skeleton bay type
of diversion at the project. Wells Project has ten powerhouse
units located directly under the spillway bays, in the spillway
section of the dam. There are no plans for construction of
future generating units in this project and it appears that
major construction would be necessary to provide any form of
bypass capability here. Chief Joseph now has 16 generating
units installed with 11 more units planned in 1976-78. Chief
Joseph Dam utilizes a penstock intake to the turbine generator
units, so the diversion through empty bays would be consider-
ably different than the skeleton bays previously described.
-------
75
However, it does appear that bypass through these empty units
is a definite possibility. At Grand Coulee, there are plans
for three additional generating units plus a third powerhouse
with nine additional units. No information is available about
bypass capability.
Figure 17 shows the present potential capacity to pass
flow through structures other than the spillway at all the
projects, assuming the turbine-generator units were fully
loaded. At each of the projects the capacity of the diversion
through the skeleton bays was assumed to be roughly equivalent
to the hydraulic capacity of the turbine generator units if
they were spinning under full load* Also shown, for comparison,
are approximate regulated river flows in an average and high
water year after Libby, Mica, and Dworshak reservoirs are
operational. The low water year (1944) would produce no spills.
The flows shown do not account for any effects of some addi-
tional manipulation of storage to provide additional control
of spills.
It is apparent from Figure 17 that even with the maximum
amount of present bypass capability installed there would still
be significant spills at many dams. The hydraulic imbalance
of the total system, as shown on the figure, is one of the
primary causes. There is relatively little useable storage
available in most of the "run-of-the-river" projects in the
Columbia and Snake Rivers. Consequently, flows past one dam
are roughly equal to the releases at the upstream project plus
any tributary inflow between the two. So, if a project with a
total hydraulic capacity of 450,000 cfs (such as John Day)
were passing that flow through its turbines, Bonneville, down-
stream with a turbine capacity of only 250,000 cfs, would be
forced to spill 200,000 cfs or more. This points to the
need for some form of additional bypass facilities at many of
the projects. This extra bypass facility could take one of
several different forms, depending on the needs at the parti-
cular projects, but would be required to pass flows in excess
of turbine capacities without entraining atmospheric gasses
in the water. A few of the ideas which have been proposed
to date will be briefly described.
One such facility would be some form of a low level regu-
lating outlet. This type of outlet would be a hole through
the dam at a low level, as simple as possible, and with a gated
structure to allow flow through the outlet to be regulated
-------
Capacity
-------
77
according to the need for passing flood flows (see Figure 18).
The outlets could be designed to pass all flows in excess of
turbine capacity in a high flow year, say the one in ten year
flood. If the downstream outlet discharges far enough below
the water level in the stilling basin, the turbulent, energy
dissipating hydraulic jump will be completely submerged and
will not entrain atmospheric gasses. If sufficient depth is
not available for complete submergence, another means of dissi-
pating the energy in the bypassed flow may be required. One
such method could be the installation of a multi-orificed
gate in the outlet conduit. Energy would then be dissipated
before the flow arrived at the outlet, and turbulence and
air entrainment in the stilling basin would be minimized.
Another form of bypass facility would be a long, stepped,
open channel around the dam, A third method of bypassing
flows is closely related to the skeleton bay diversion and
would involve special turbine design to allow them to be ad-
justed to operate "inefficiently." Then, during periods of
high flows and low power demand the units could be adjusted
to pass flows up to their maximum hydraulic capacity but
with reduced power production. Another method of accomplish-
ing this would be to install slotted gates in operational tur-
bine bays. The gates would dissipate much of the energy in
the flow so that correspondingly less power would be trans-
ferred to the generator. Studies along this line are in
progress at the Corps of Engineers District offices and the
North Pacific Division Headquarters, but considerably more
work needs to be done before feasible designs will be ready.
In reviewing generator installation schedules and
turbine-generator construction methods, one additional
operational-structural measure for reduction of nitrogen
problems should be considered. Normally, when several gen-
erator units are to be installed at a project, because of
construction and supply scheduling demands, they are built up
in a "stair-step" sequence. This means that at any given time
different phases of construction are underway at each unit
and each is very likely at a different stage of completion.
With this method, all the turbine bays need to be dewatered
at the same time and are unavailable for diversion of flows
until the entire set of new units is nearly complete. A
couple of possible solutions to this problem appear to be
available. One might be the complete installation of one unit
at a time, leaving the other skeleton bays open and available
for diversion of flow. Another possible solution would be the
"capped-unit" type of construction mentioned earlier. With
-------
Tailwater El. 638
Figure 18 Proposed Configuration of a Low-Level Regulating Outlet
(Lower Granite Spillway)
1/ Corps of Engineers, North Pacific Division
>
Lov/ Level Outlet
merjzt/ /-/tj dro
El.
Deck El. 751
Pool El
-------
79
that type of construction, all of the units could be worked
on simultaneously, while at the same time flows could be
bypassed through the lower cavity.
Another type of structural solution which has been given
some consideration is a modification of the spillway and still-
ing basin combination. (Anonymous, 1971C). The basic function
of the stilling basin is to provide a structure having the
necessary length and depth for energy dissipation (in the form
of a hydraulic jump) for all required flows, including the
major flood flows. Some available evidence indicates that
certain stilling basins tend to produce a lesser supersatura-
tion level than others do. In order to more closely simulate
the characteristics of those stilling basins having the lesser
and desirable levels of nitrogen saturation, high releases
could be concentrated through one or more of the spillway
bays. However, since this drastic alteration in spill patterns
could seriously jeopardize upstream migrant fish collection
and passage, this solution is questionable. To lessen the
depth by raising the stilling basin floor could jeopardize
the integrity of the structure by providing inadequate depth
to dissipate energy for major flood flows.
Recent studies by the Corps of Engineers indicate that
it may be possible to modify the spillway structure to cause
spillway flows to plunge to a lesser depth in the stilling
basin. The concept involves a lip on the downstream face of
the spillway structure, which directs small to moderate con-
centrations of flow along the surface of the tailwater (see
Figure 19). Additional studies plus prototype tests will be
required before this scheme can be considered feasible,,
However, it does oifer an attractive possibility for safe
bypass of some excess flow at many existing projects where
provision of other types of bypass structures will be extremely
difficult.
Other Methods of Controlling Nitrogen
Additional Reservoir Storage
In this section of the report, the physical possibility of
eliminating flood flows over spillways will be discussed,
regardless of any present legal or operational restrictions.
-------
Deck El. 751
Poo! El. 733
El. 681
10'-2
Spillway Deflector
Tailwater El. 638
Figure JL9 Proposed Configuration of Spillway Deflector (Lower Granite Spillway) ^
.1/ Corps of Engineers, North Pacific Division
-------
81
One method of reducing spills, of course, would be the
establishment, of more complete control of flood run-off in
the Columbia and Snake River system. The flow of the lower
Columbia River is now regulated by upstream storage reservoirs
located principally in headwaters areas of the Snake River
and in Canada. At present, there is about 22,000,000 acre-
feet of active storage,, This will soon be increased to
nearly 41,000,000 acre-feet when Libby, Mica, and Dworshak
Reservoirs, now under construction, are completed (about
1973)c Some preliminary computations were made by personnel
of the Corps of Engineers to determine the possible effect
that some additional storage could have on reducing spills
in the lower river. For this analysis it must be noted that
a reservoir at any particular point can influence downstream
flows only to the extent that the tributary on which the res-
ervoir is located actually contributes flows to the downstream
system. Figure 20 shows the effects on the 1980 condition of
additional storage at a few projects. For these calculations,
approximately 3 million acre-feet of storage was assumed to
be constructed on the upper Columbia with about 2.4 million
being actually available for flood control, and 5.5 million
additional acre-feet of storage would be constructed on the
Snake River with approximately 3.8 million being available for
flood control. None of this storage is presently authorized
and so should be considered highly speculative at this time.
However, the analysis does serve to illustrate the effect on
stream flows of upstream storage of these rough magnitudes.
Columbia River Treaty Storage
Under terms of the Columbia River Treaty between Canada
and the United States (Anonymous, date unknown), 8.45 million
acre-feet of Canadian storage is routinely available to the
United States for use in their flood control operating plans.
Each year by May 1, three Canadian Reservoirs -- Arrow, Duncan
and Mica -- are to be drawn down sufficiently to provide this
8.45 million acre-feet of available storage if required for
flood control in the U. S. The treaty also requires that
Canada operate any additional storage in the Columbia Basin
on an "on-call" basis when required by the United States for
the purpose of controlling exceptionally high floods. This
"on-call" storage could make up to 20,500,000 acre-feet
(including the 8,450,000 acre-feet routinely available) of
Canadian storage available for flood control in the U. S.
For use of the "on-call" storage by the United States, the
-------
(" t' IPaq, Ftaij^,
•' ';. i;!11.'/^,:;: Wampum £
Pr /est Ra-pids
I-., I1'''f1 *!
'I l'iL
'll^i
' •! < 1111
i ¦ -
11 .
«j jo
I; 1
0
M '
'1 1!''!1! i t. ;, .1 i1 i1 ¦:1:'! 11 ¦ i, i m 1 ¦1 ¦ ¦VT
¦¦ !''!m "• '.'.I'.i i1:!'!" ;•! .
¦ , . ;;r.' i.
•!.' .i!*'' !1 i'1 1: ¦ ;t 'I' ;|| Ji.!1' !¦ .!¦
:• 'i"!1'! :;n iJ'i:!;:;-:!!1
ll111, •1 t J .
I M
"i'ii
j111 \\ •:.i,;11 ¦,|*,j 1. j111.j 11.j!|!![ ' !j':!!I ¦| ¦. i I•«jj1'
I i ¦: i1' I. 1 1 :
:i 1 '
tS JO1/ // ( jyj/ ( tf Jdjf
v' 4 »' !' I! • i 1 ¦.'' 1 '''
I I' 11 ' 1' 11
• ll'IIJ
~ *1 ill"'
July i . 11 ; I
i! !'<
iiiil:
J - . " 3(x\
III •
['/' \fiJ ¦ '• I'J.'iJf i]'' r ''I I, ; |
.!rl*Lir 11h'¦ ¦""y.5111i"i HO* i.
!!i!^T:i^ ^iH iiiU's ilir "i'ii! Lower Granifc
"lilM'i
Wr r *J
. r' '
:1 >' ¦ v
ll-'^MKiiViiiifailllK
" •' "r i' ; ¦"!' 1 ^ i;r!."I
Flgur* Efface of addition*! upatraaa atoraga, high flow yaar <1956)
ToM £/ouj
pQ*Sinj pfQj «c/
Mgdrautu, C/jpQC'Jy
of fvrSmcS
— • y//
Ficut +t\r*ogk
ivrkints
C+Q' f*o
-------
!r
!>¦'
!i"i I'
!': : 11!'f"1 ¦.11- III'-';!'! i|i;|;: ih:i
lilL
I..'] I
* I 1 r I ! | I I J \ | I
ii''!ili'!
Stirirqt
n: ¦.
'![
: i -1
• 'jj j V\/a n a. p u m ^
' ii'li.linjililjii! Pri«s+ Ra pids
i'11 f; 1 :¦ .¦¦¦'I'ljl;:! ;
ftmjf, juac ¦ r y«
i ' . 'l.m' 'I'l; ¦ .iiii;,
¦f:!i #!;;
¦¦ l!' "IiT'l"fi
'"'I'liliiS;!
i! i" r
ii
ul'll '
•'T1
i 'i
hli'iiiii i
-i-
i
rr
V.i'
¦¦ !!!W|!li!S!iSi
lli'l
ilii!
B
onnt i/i
//<
hi:!:
of ctdJ/Jional
i'' Storage !
I y
L_o_w
l" ~
I-
/ - . tS
/f
-' i' *
ir j* i
/r
! "J _ Af"->J i
Jwnt j
j*ij
, • ":i
,: —j : ""1
* ~
* '
To+o/ £touj
pats/nj />roy «c/
Pjc/reu/n. CapQCtJj
of ivr-Ltnci
Ptou» through
ivf&iics
¦ » y//
Lcgcnc/
Figure Effect of additional upaeraaa atorage, average flow year {Coluabla-1952, Snak.e-1932)
-------
84
U. S. must pny Canada a sum of $1,875,000 each of the first
four times the additional storage is called upon. The U. S.
must also deliver to Canada electric power equal to the hydro-
electric power lost by Canada as a result of operations to
provide the additional storage. This arrangement is to
continue for a period of 60 years after the date of ratifica-
tion of the treaty '(September, 1964). However, other provisions
of the treaty require that this "on-call" storage shall not
be requested unless flood forecasts predict a peak discharge
in excess of 600,000 cfs at The Dalles, Oregon (regulated by
all available storage in the U. S. and Canada).
The presentation in previous figures of the hydrographs
fdr floods expected in a high flow year (1956) were based on
computer studies in which it was assumed that a total of
7,000,000 acre-feet of storage would be available in Mica
Reservoir at the start of the flood season. This means that
about 5,000,000 acre-feet of this would have been part of
the "on-call" storage. In the analysis of the average year
flood (1952) it was assumed that "on-call" storage would not
be required.
Although there is 10,000,000 acre-feet of storage avail-
able at Mica in addition to the 2,000,000 acre-feet required
for flood control, not all of this should be considered a
potential additional source for spill reduction during the
critical periods of the flood runoff. First, winter power
requirements usually result in Mica being drawn down to an
extent such that more storage than the 2,000,000 acre-feet
is available, at the beginning of the flood season. For
example, the "average year" hydrographs shown on Figure 10b(2)
reflects the use of 5,760,000 acre-feet of storage space at
Mica. Thus any special "on-call" space at Mica designated
for spill reduction is likely to be considered less than the
total 10,000,000 acre-feet. Second, due to minimum release
restrictions at Grand Coulee there is a limit to which lower
river flows can be reduced during the height of the flood
runoff. Finally, if a special "on-call" system were imple-
mented, this storage could not be made available in every
large runoff year, statistically speaking, since not all
large runoffs can be forecasted far enough in advance to get
the Mica pool drawn down far enough to make the additional
"on-call" storage available during the flood season.
It is very difficult to determine the actual effect on
downstream flows of using or not using the "on-call" storage
-------
85
available In Canada. This is because of the irregular sequence
of inflows, storage releases, and buildup of flows downstream,
However, results of one study conducted by the Corps of Engi-
neers, using the system as it would be regulated after 1975,
and based on a relatively high runoff year (1951) do give an
indication of the possible magnitude of the additional down-
stream regulation. In that study, downstream flows under an
integrated system operation were calculated for two separate
cases. In the first case, the storage available in the
Canadian projects at the beginning of the flood season was
approximately 7,000,000 acre-feet. In the second case, the
"on-call" storage was utilized and approximately 14,000,000
acre-feet were assumed to be available in the Canadian projects
prior to the flood season. The results showed essentially no
difference in streamflow between the two cases until approxi-
mately May 15. During the end of May and the first part of
July, the use of the additional "on-call" storage reduced
streamflows at The Dalles by about 40,000 cfs. Interestingly,
the greatest effect of the "on-call" storage occurred later
in the year. In July, the use of the "on-call" storage was
found to reduce the flow by some 100,000 - 150,000 cfs.
This use of "on-call" storage would be subject to the
cost limitations cited previously. Also, the Columbia River
Treaty would undoubtedly have to be amended. However, the
fact that the additional storage could provide substantial
extra control of downstream river flows suggests that it should
be given further consideration and that additional studies
should be directed to determine more precisely the actual
level of downstream control which could be obtained, the true
cost of the operation in a particular year, and the procedural
difficulties which would be involved in revising the treaty
and preparing flood control plans for its eventual use„
A last method of controlling spills which is physically
possible but which may have serious operational, structural
and possibly legal concerns operating against it, would be
the provision of complete bypass capability0 This measure
would require that each dam in the system have some type of
bypass facility which could, without entraining excess nitro-
gen, pass any flows in excess of flows which could safely be
passed through turbine generator units, skeleton bays, and
navigation locks. The capacity of this diversion facility
could be based on a flood selected to have such a frequency
(say, the once in ten year event) that recurrence of nitrogen
-------
problems that often would not result in undue damage to the
fishery resource.
Effects of Future System Operation
As population and industry continue to expand and power
requirements continue to increase, the systems used for
generating power will grow and change. The following dis-
cussion of changes in the use of reservoir storage for power
generation was developed by power planning agencies in the
Pacific-Northwest (Anonymous, 1970):
"The pattern of storage regulation for hydro-
electric power production will change as the Pacific
Northwest integrated power system progresses from a
hydro base to a thermal base. During the initial
stage all available power storage will be used during
the adverse water years to produce the maximum
possible prime power. This will establish the maxi-
mum firm load which can be carried by the system,
and therefore, in years of better than minimum flow,
the available storage will not be fully used for
power generation. During the second stage of de-
velopment, thermal generation will grow rapidly,
and installations at main river hydro plants will
be expanded. All available energy, including much
that was formerly spilled, will become usable for
replacing thermal electric energy. As much storage
will be withdrawn during the winter season as can
be replaced with forecasted flood season flows.
As a result, the average annual use of storage will
be greater than in the initial stage. During periods
of adverse streamflow, all power storage necessary
to meet firm loads will be used as it would be in
the initial stage.
"In the ultimate stage of development, reser-
voirs will [could] \f be maintained at a relatively
high level to provide full plant peaking capability
until the January peak load has occurred. Follow-
ing the annual peak load, storage will be with-
drawn, on a forecast basis, to generate hydro energy
to replace thermal energy and to prevent subsequent
spill during the flood runoff season. The change
\! Preferred wording by BPA
-------
87
from one stage to the next will be gradual, and
the length of each period will depend on the rate
of load growth, the rate of adding new storage and
thermal generation, and the magnitude of the ulti-
mate hydro-installed capability. The common ob-
jective in all three periods is to reduce the spill
of water during the high runoff period by storing
flows in excess of downstream plant hydraulic
capabilities and thus convert potential spill into
a usable commodity."
A major area of concern related not only to nitrogen
supersaturation and the fishery resource but also to other
water uses of the Columbia River system, such as recrea-
tion, navigation, and wildlife habitat, is the possible
future effects of the increased use of hydro projects on the
river for "power peaking." "Power peaking" is the term
used to refer to power generation operations designed to
meet the diurnal fluctuation in power demands. Figure 21
shows the fluctuation in load during the hours of a typical
day in the Pacific Northwest. In the diagram and the dis-
cussions which follow, the firm load represents that portion
of the demand for which firm contracts for power have been
written between the power user (a municipality or industry)
and the power marketing agency. The secondary load is
additional load, usually industrial, for which there are
no firm contracts but which could be served providing there
is enough generating capability available. Municipal power
use is the primary factor responsible for the wide variations
in hourly loads, indistrial loads normally tend to be rela-
tively stable during a daily cycle.
Table 3 shows projected future electric power requirements
in the Columbia-North Pacific region. The "load factor" is
the ratio of the average load during a period to the peak
load occurring in that period and expressed as a percentage.
It may be noted that although there is a 470 percent increase
in average megawatts of demand from 1980 to 2020, the load
factor through this same period remains relatively constant.
Table 3 shows that in 1980 the peak power demand will be
approximately 12,400 megawatts above the average load; in
2020 the need for peak power will be approximately 66,600
megawatts above the average power load.
-------
9
8
2
1 "
0 -I
00 03 06 09 12 15 18 21 24
TIME, HOURS
Figure ^ Daily Load Shape, Federal System 1 /
1/ Reference: Anonymous, 1971C
-------
89
TABLE 3
1/
1965
1980
2000
2020
Population (millions)
5.9
7.3
9.7
12.7
Ratio: Pop./Dom. Cust.
3.1/1
2.8/1
Domestic Customers 1
KWH Use per Dom. Cust.
Total Domestic Use GWH —
,878,814
11,011
20,687
2,607,000
17,700
46,144
Ratio: Pop./Com. Cust.
24.5/1
22.5/1
Commercial Customers
KWH Use per Com. Cust.
Total Com. Cust. GWH
239,883
36,607
8,781
325,000
67,200
21,840
Industrial Customers
Industrial KWH Use per Capita
Total Industrial Use GWH
6,486
5,936
34,346
99,000
Irrigation GUH
2,421
6,593
Other GWH
1,090
2,259
Total Sales GWH
Losses
Total Requirements GWH
67,325
7,110
74,435
175,836
17,364
193,200
479,000
1,096,000
Per Capita KWH Requirements
12,676
26,500
49,400
86,300
Peak-megawatts
13,068
34,400
84,800
191,700
Average-Megawatts
8,497
22,000
54,660
125,060
Load Factor—Percent
65.0
64.0
64.4
65.0
Electric Power estimates for the region with alternate higher popula-
tion projections in the Puget Sound and Willamette subregions would
be as follows:
2000 2020
Total Requirements
Million KWH 512,000 1,286,000
Peak-Megawatts 91,300 229,400
1/ Reference: Anonymous, 1970
2/ All kilowatt-hour data refers to annual use.
-------
90
Hydroelectric plants at dams on the Columbia and Snake
Rivers have the unique ability to start quickly and change power
output rapidly. This makes them especially suitable for carry-
ing peak loads and supplying spinning reserves. For this reason,
present plans for power development in the Columbia-North Pacific
region call for the expansion of hydroelectric generating
facilities to be used to generate greater and greater portions
of the peak power demands, and thereby operate at least lower
load factors.
Both the Libby and Dworshak power plants have been designed
to operate ultimately at an annual capacity factor (see Glossary)
of approximately 20 percent. In other words, they will be able
to generate peak power five times greater than the average load
carried. At the Columbia River and lower Snake run-of-river
projects, facilities are also being included in the initial
phase of development for future installation of additional
peaking units. However, in the interest of navigation and
other water uses, the ultimate annual capacity factor of these
plants has been limited to about 40 percent. The amount of
base load which will be carried by these hydroelectric plants
will thus decrease in the future, leaving thermal generating
facilities to carry more and more of the base load, until
eventually thermal generating facilities will be carrying
nearly all the base load. Table 4 lists the projected thermal
capacity which will be installed in future years and Table 5
is an installation schedule for generating units at hydro
projects in the Pacific Northwest region. These tables are
under continuous revision as power demand forecasts are revised
through the years.
Because, generally, thermal plants are more expensive to
operate than hydroelectric plants, the operating procedure
for the future will be as follows: Firm load plus secondary
energy loads will be generated by hydroelectric power resources
until the full load is met—thermal power will be completely
shut do™ if necessary to reduce spills. The foregoing state-
ment applies mainly to seasonal or relatively long-term shifts
in power generation. The statement must be tempered with the
availability of loads, with the load factoring capability
of hydroelectric plants, and with the cost of thermal replace-
ment energy.
As the total system load increases, load factoring will
increase at the hydro plants, with consequently increasing
amounts of daily spills during the off-peak periods. This is
one of the primary effects of power-peaking operations.
-------
91
TABLE ^INSTALLATION SCHEDULE FOR THERMAL PROJECTS 1/ and 5/
Plant
Type
of
Rating 2/
Date
Sponsor h/
No.
Name
Fuel
Unit
Scheduled 3/
1
Centralia
Coal
1
2
700
700
Sept.
Sept.
1971
1972
PP&L Co. &
WWP Co.
2
Trojan
Nuclear
1
1,106
Sept.
197k
PGE Co.
3
Jim Bridger
Coal
2
3
500
$00
Sept.
Sept.
1975
1976
PP&L Co.
PP&L Co.
li
Hanford §2
Nuclear
1,100
Sept.
1977
WPPSS
5
—
Thermal
1,100
Sept.
1978
Private
6
—
Thermal
1,100
Sept.
1979
Private
7
—
Thermal
1,100
Sept.
1980
Public
8
—
Thermal
1,100
Apr.
1982
Private
9
—
Thermal
1,100
Apr.
1983
Private
10
• •
Thermal
1,200
July
198^
Public
1/ All information for plants £ through 10 is tentative.
?/ Size of scire units may vary from that forecast if a different type of
fuel and/or location i3 selected. Plants 6 through 10 bright be as
large as 1,500 megavatts.
3/ Schedules rcay be rearranged between projects depending on needs of
sponsoring utilities.
h/ Sponsors for plants 8 through 10 have been tentatively identified only as
to type of utility according to present estimates of energy requirements
5/ Prepared by Bonneville Power Administration, Branch of Power Resources,
December 29, 1970.
-------
92
TABLE 5 INSTALLATION SCHEDULE FOft HYDRO PROJECTS 1/
(See footnotes on page 94 )
Project 2/
-Rocky Reach Additions
(Chelan Co. FUD)
fohn Dsy*
"The Dalles*
)worshak*
¦Grand Coulee
Pump Generators'"'
3rd Powerp] ant»-
Libby*
Initial
Unit
Nameplate
Date
of
No.
Mer.awatts 3/
Operation h/
8
125.4
July
1971
9
125.4
Sept.
1971
10
125.4
Oct.
1971
11
125.4
Nov.
1971
15
135
Aug.
1971
16
135
Oct.
1971
17, 18
135 ea.
Oct.
1988
19, 20
135 ea.
Dec.
1968
15
86
Aug.
1972
16
86
Nov.
1972
17
86
Feb.
1973
18
86
May
1973
19
86
Aug.
1973
20
86
Nov.
1973
21
86
Feb.
1974
22
86
May
1974
1
90
Nov.
1972
2
90
Feb.
1973
3
220
Kay
1973
-4
220
Aug.
1985
5
220
Oct.
1985
6
220
Dec.
1985
P/G-7
50
Apr.
1973
P/G-8
50
June
1973
P/G-9—P/G-12
50 ea.
Sept.
1989
G-19
600
Feb.
1974
G-20
600
Aug.
1974
G-21
600
Feb.
1975
G-22
600
Sept.
1979
G-23
600
Mar.
1980
G-24
600
Sept.
1980
G-25 W
600
Sept.
1988
G-26 5/
600
Mar.
19S9
G-27 tf
600
Sept.
1989
1
105
Jul.
1975
2
105
Oct.
1975
3
105
Jan
1976
4
105
Apr.
1976
UC
UC
UC
UC
UC
UC
F
F
UC
UC
UC
UC
UC
UC
UC
UC
UC
UC
UC
F
F
F
UC
UC
F
UC
UC
UC
UC
UC
UC
F
F
F
UC
UC
UC
UC
-------
TABLE 5INSTALLATION SCHEDULE FCR HYDRO HiOJECTS 1/
(Continued)
Initial
Unit
Nameplate
Date
of
Project 2/
No.
Megawatts 3/
Operation h/
Ice Harbor*
k
111
Feb.
1975
5
111
May
1975
6
111
Aug.
1975
Lower Granite*
1
135
Apr.
1975
2
135
Apr.
1975
3
135
Apr.
1975
h
135
Feb.
1983
5
135
May
1983
6
135
Aug.
1983
Lost Creek#
1
2ii.5
Apr.
1976
2
2U.5
June
1976
Teton#
1
10
May
1976
2
10
June
1976
3
10
June
1978
Chief Joseph*
17
95
Nov.
1976
18
95
Feb.
1977
19
95
May
1977
20
95
Aug.
1977
21
95
Nov.
1977
22
95
Jan.
1978
23
95
Mar.
1978
2h
95
May
1978
25
95
July
1978
26
95
Sept.
1978
27
95
Nov.
1978
Bonneville
11
5b
Feb.
1978 7/
2nd Powerplant*
12
%
May
1978
13
Sh
Aug.
1978
Hi
5U
Nov.
1978
15
51
Feb.
1979
16
5U
May
1979
Little Goose*
U
135
Feb.
1978
5
135
May'
1978
6
135
Aug.
1978
Asotin*
1.
2
135 ea.
Apr.
1980
Guffey
1,
2
, 3, b 21.25 ea.
July
1981
Lower Scriver
1.
2
30 ea.
May
1982
3,
it
30 ea.
July
1982
-------
94
1/
TABLE 5 INSTALLATION SCHEDULE FOR 1TCDRO PROJECTS -
(Continued)
Includes projects under construction, authorized, recom-
mended arid proposed (potential) „
Ownership is Federal unless otherwise shown,,
Manufacturer's rating and does not necessarily represent
maximum capability.
Unit schedules for installation subsequent to 1980 are
tentative and are subject to change depending on pro-
jected loads and plans of constructing entities.
UC - Under construction
P - Planned for early installation
F - Future
Authorization required for these units and for units
G-28, 29, and 30.
Based upon a site selection having a construction period
of five years.
Authorized for Federal Construction
-------
95
Table 6 shows a load-resource analysis for the Columbia-
North PaHfic region, assuming maximum development of the
hydro system. Under LhJ :i |>l;in, hy the year 2000, cssuniti al ]y
all of Lhe hydroelectric resources will have been developed.
The table also gives an indication of the magnitude of the
maximum power-peaking that could be required from hydroelec-
tric resources. For example, in the year 2020 the hydroelec-
tric resources would carry 15,000 megawatts of the average
energy load but would be peaked up to 56,000 megawatts to
meet peak power demands, while at the same time the thermal
resources would be held relatively constant. This could mean
a huge amount of daily spilling as hydroelectric power-
generating facilities are dropped off-line at night during
the off-peak periods.
The basic assumption used in preparing Table 6 was that
the future thermal plants would be designed to operate most
efficiently on a continual basis. It is generally assumed
that these pLants would operate at an 85 percent plant factor'
(see GLOSSARY). The 15 percent reserve would allow for forced
outages, maintenance, etc. Their full, rated capacity would
be dependable for meeting critical period peaking requiremants.
It was also assumed that when sufficient water is available
to the hydro-electric system, energy would be produced
there for replacement of thermal power generation and the
thermal plants would be shut down in order of decreasing
energy cost (fuel costs). With no load factoring at the
thermal plants, then, it can be seen that the hydroelectric
generating plants have to be, in the future, varied greatly
on a diurnal cycle. This would result in a diurnal cycle of
spilling and no-spill during the high flow season (see
Figure 22a).
Existing technology in thermal-electric generating facili-
ties, however, indicates that thermal plants may be load
factored to some extent. A limit on load factoring of thermal
plants would probably be at a maximum rate of about 5 percent
per minute. This amount of flexibility could significantly
assist the effort to reduce diurnal spilling at hydroelectric
projects during the high water season. If this capability
is to be utilized, complex scheduling and coordination of
fuel supply contracts, refueling operations, maintenance, and
repairs will be required. Figure 22a shows pictorially the
way in which the diurnal spills could fluctuate during the high
flow season under a normal peaking operation in the future.
Figure 22b shows the type of operation which could result with
some load factoring at the thermal plants. Available pondage
and storage are assumed used in the same fashion for both
Figures 22a and 22b.
-------
96
Table 6
I.ami Jtc'imirci' An.'i I y» I» : Plim A CcO unil) 111-NorLh Pacific Rep,ion —1
?/ 5/
OBERK — Population Estimates Maximum Population Estimates —
Item 19804/ 2000 2020 1980^ 2000 2020
Energy (Thousand Megawatts average)
18.2 54.7^ 125.ll' 18.2 58. & 146.4'
Reserves 0.5*^ 6.9l' 19. 32/ 0.5^' 3.2Z/ J3J.-'
Total Load 18.7 61.6 144.4 18.7 61.6 169.9
Hydro Resources
Thermal Resources
Thermal Imports
Total Resources
11.9 15.5
6.3 38.1
0.5 8.0
18.7 61.6
15.5 11.9
116.9 6.3
12.0 0.5
144.4 18.7
15.5 15.5
38.5 142.4
8.0 12.0
62.0 169.9
Peaking Capability (Thousand Megawatts)
Load 30.0 84.g£/ 191.7-' 30.0 91.3^' 229.41'
Reserves 3.8 10.2 23.1 3.8 11.0 27.5
Total Load 33.8 95.0 214.8 33.8 102.3 256.9
Hydro Resources
Base Load Thermal
Thermal Imports
Peaking Resources —'
Total Resources
26.3
56.0
56.0
26.3
56.0
56.0
7.3
38.1
116.7
7.3
38.5
142.4
0.2
R.Q
12.0
0.2
8.0
12.0
0.0
0.0 .
30.1
0.0
0.0
46.5
33.8
102.li
214.8
33.8
102.5.12/
256.9
1/ Maximum hydro system.
If Based on Columbia-North Pacific population forecasts prepared by 0BE-ERS.
3/ Same population forecasts as above for 10 of the 12 subregions; population forecast
for Subregions 9 and 11 from the respective Type 2 studies.
4J Data derived from Pacific Northwest Utility Conference Committee West Group Foreca
Estimates.
5/ From table 22 (Anonymous, 1970)
6/ Energy reserves for 1980 based on one-half year's utility load growth.
Tj Fifteen percent of the required thermal capability (including imported thermal).
J8/ Based on 12 percent of load. Reserves for 1980 computed using Pacific Northwest
Coordination Agreement probability analysis formula.
9/ Pumped storage, gas turbines, or other peaking resources.
10/ Surplus capability resulting from import of thermal energy.
11/ Reference: Anonymous, 1970.
-------
91
Figure 22a. Load Factoring at Hydro Plants Only
uf vo/'eo £ f-'ovj
c^uns&'e'ir
V
. u * ~ jr j
""b
$
?
-------
98
Proponed |>uiiip-.«:l nni|;i' ojutjiI I ohm (sec d I ;j cu:;;> J on of
pumped .sl'or.ij'.i' in tin.- APPENDIX) ;it many liydro projects offers1
another possibility for reducing d;iily spills. At the
presently projected rate of power development, supplemental
peaking capacity beyond that available by development of
conventional hydro-projects will probably not be required
until after 1990. Studies are now being conducted to deter-
mine when and how pumped storage will best fit into the
regional load projections. It is evident that pumped storage
offers considerable promise as a source of future peaking
capacity and spill reduction.
The above comments have all been based on load and power
generation forecasts on a region-wide basis. At this time,
there are 110 detailed studies available which indicate the
character of seasonal and diurnal spills in future years.
However, computer proftmms and basic data are available to
perform tins type of .incilysis and these studies should be
conducted in the near future. Objectives of such a study
should be to determine how much spilling may occur in the futu
the detailed effects of power peaking and pumped storage, and
the feasibility of using load factoring at the thermal plants
to help reduce seasonal and diurnal spilling at the hydro-
electric projects.
Needs for Further Study
Throughout the report, gaps in data and needs for addi-
tional study have been mentioned as each topic was discussed.
The following listing summarizes the needs for additional work
related to engineering aspects of the nitrogen problem.
1. Detailed computer studies should be oriented toward
predicting levels of spills in future years and unde
variations in the system operation discussed in
this chapter.
2. Studies of estimated costs of the many alternatives
for controlling nitrogen should be made to provide
bases for decisions on implementation.
3. Studies on turbine-generator design should be con-
tinued toward development of units which can pass
large quantities of water at low load without
damaging effects to fish or structures.
-------
99
4. Additional studies should be performed to develop
non air-entraining spillway stilling basin designs.
5. An inventory of bypass capability at each dam on
the Columbia and Snake Rivers should be made.
6. The potential effects of Canadian Treaty Storage
should be investigated further.
7. Methods of reducing the detrimental effects of
power peaking should be studied further. These
methods include load factoring at proposed thermal
plants and operation of future pump-storage projects.
8. Better correlations between spillway flows and
nitrogen levels are needed.
-------
100
BIBLIOGRAPHY
* Anonymous, "Flood Control Operating Plan for Columbia River
Storage," International Task Force on Flood Control
Operating Plan, U„ S0 Army Corps of Engineers, North
Pacific Division, and Bonneville Power Administration,
Portland, Oregon, August 12, 1968A.
Anonymous, "interim Flood Control Operating Plan for Duncan
and Arrow Reservoirs," Bonneville Power Administration,
Portland, Oregon, November 12, 1968B.
* Anonymous, "Columbia River Treaty Documents," U. S. Department
of the Interior, (publication place and date unknown).
* Anonymous, "Appendix XV, Electric Power," Columbia-North
Pacific Region Comprehensive Framework Study of Water
and Related Lands, Pacific Northwest River Basins
Commission, Vancouver, Washington, October 1970.
Anonymous, "Columbia River Thermal Effects Study Vol. I
Biological Effects Studies," Environmental Protection
Agency, Pacific Northwest Region, Portland, Oregon,
January 1971A.
Anonymous, "Review of Power Planning in the Pacific Northwest,
Calendar Year 1970," Pacific Northwest River Basins
Commission, Power Planning Committee, Vancouver, Wash-
ington, January 1971B.
* Anonymous, "Conference Proceedings, Nitrogen Studies Meet-
ing, Corps of Engineers with Northwest Fishery Agencies
and Environmental Protection Agency," U„ S„ Army Corps
of Engineers, North Pacific Division, Portland,
Oregon, February 17, 1971C.
Anonymous, "Nitrogen Supersaturation Problem, Public Meeting ~
March 23, 1971," U. S„ Army Corps of Engineers, North
Pacific Division, Portland, Oregon, March 1971D.
* Bexningen, Kirk T., and Wesley J. Ebel, "Effect of John Day
Dam on Dissolved Nitrogen Concentrations and Salmon in
the Columbia River," Trans, of Amer. Fisheries Soc.,
Vol. 99, No. 4. October 1970.
* Literature cited.
-------
101
* BeJninp.tMi, Kjrk T. and Wesley J. Kbel, "Dissolved Nitrogen,
Dissolved Oxygen, and Related Water Temperatures in
the Columbia anil ],ower Snake Rivers, 1965-69," Data
Report No. 56, National Marine Fisheries Service,
Seattle, Washington, April 1971.
Bishai, H. M. "The Effect of Gas Content of Water on Larval
and Young Fish," Z. Wiss. Zool. 163:37-64, 1960.
* Bouck, G., G. Chapman, P0 Schneider, D. Stevens, and J.
Jacobson, "Initial Studies of Temperature Requirements
of Adult Sockeye Salmon (Oncorhynchus nerka). Adult
Coho Salmon (Oncorhynchus kisutch). and Thermal-Chemical
Requirements of Juvenile Steelhead Trout (Salmo
gairdneri) in the Columbia River," Federal Water
Quality Administration unpublished report, October 1970A.
* Bouck, Gerald R., Gary A. Chapman, Phillip W, Schneider, Jr.
and Donald G. Stevens, "Observations on Gas Bubble
Disease in Adult Columbia River Sockeyc Salmon
(Oncorhynchus nerka)," Pacific Northwest Laboratory,
Federal Water Quality Administration, Corvallis, Oregon,
June 1970B (unpublished manuscript).
Coutant, Charles C., "Exploratory Studies of the Interactions
of Gas Supersaturation and Temperature on Mortality
of Juvenile Salmonids," Battelle Memorial Institute,
Pacific Northwest Laboratories, Richland, Washington,
(unpublished report), September 28, 1970.
* Coutant, Charles C., and Robert G. Genoway, "Final Report
on an Exploratory Study of Interaction of Increased
Temperature and Nitrogen Supersaturation on Mortality
of Adult Salmonids." A Report to U. S„ Bur. of Com-
mercial Fisheries, Seattle, Washington. Battelle
Memorial Institute, Pacific Northwest Laboratories,
Richland, Washington, November 28, 1968.
* Craig, H. and R. F. Weiss, "Dissolved Gas Saturation
Anomalies and Excess Helium in the Ocean," Earth and
Planetary Science Letters. Vol. 10, No. 3, Febru-
ary 1971.
* Ebel, Wesley, "Effects of Transportation on Survival and
Homing Ability of Salmonids from the Snake River,"
Bureau of Commercial Fisheries, Biological Lab, Progress
Report, 12 pp„, September 1970A.
-------
102
libel, Wesley J., "Summary Report 1970 Dissolved Nitrogen
Surveys of the Columbia and Snake Rivers," prepared for
the U„ S. Army Corps of Engineers, Portland District,
Portland, Oregon, 17 pp., 1970B.
Ebel, Wesley J., "Supersaturation of Nitrogen in the Columbia
River and Its Effect on Salmon and Steelhead Trout,"
Bureau of Commercial Fisheries. Fisheries Bulletin 68.
#1, pp. 1-11, 1970C.
Ebel, Wesley J., Earl M. Dawley, and Bruce Monk, "Thermal
Tolerance of Juvenile Salmon in Relation to Nitrogen
Supersaturation," National Marine Fisheries Service
Biological Laboratory, Seattle, Washington, unpub-
lished report, September 1970D.
Egusa, Syuzo, "The Cas Disease of Fish Due to Excess of
Nitrogen," Jour. Fac. Fish. An. Husb. 2:157-183, 1959
Evans, A., and D. N, Walder, "Significance of Gas Micronuclei
in the Aetiology of Decompression Sickness," Nature.
Vol. 222, pp. 251 to 252, April 19, 1969.
Fox, J. J., "On the Coefficients of Absorption of Nitrogen
and Oxygen in Distilled Water and Seawater, and Atmos-
pheric Carbonic Acid in Seawater," Trans. Faraday
Soc. Vol. 5, pp. 68-87, 1909.
Gorham, A. M., "The Gas Bubble Disease of Fish and Its Cause,"
Bull, of the U. S. Fish Commission. Vol. 19, pp. 33 to
37, 1899.
Greenland, Donald C., "Future Operations of Mainstem Hydro-
electric Projects, Columbia and Snake Rivers," Bureau
of Commercial Fisheries, Portland, Oregon, processed
report, April 9, 1969.
Harvey, E. N., "Physical Factors in Bubble Formation,"
Ch.4, p. 90 in Decompression Sickness, ed. J. F. Fulton,
Saunders, Philadelphia, 1951.
Harvey, E. Newton, A. H. Whiteley, W. D. McElroy, D. C.
Pease, and D. K. Barnes, "Bubble Formation in Animals,"
Jour, of Cell and Comp. Physiology. Vol. 24, 1944.
Harvey, Harold H., "Supersaturation of Lake Water with a
Precaution to Hatchery Usage," Trans. Amer. Fisheries
Soc., 96:194 to 201.
-------
103
Harvey, 11 „ II,. unci A. C. Cooper, "Orlj'.ju .ttul Treaimont of a
Supurr.JlmrriCuci River tenter." ZntrcTnnC.ion.il Pacific Snlmon
Fisli Commission Progress Report: No. 9, 19 pp., 1962.
Harvey, H„ H. and S. B0 Smith, "Supersaturation of the Water
Supply and Occurrence of Gas Bubble Disease at Cultus
Lake Trout Hatchery," Con. Fish Cult.. Vol. 30. pp. 39
to 46, 1961.
Hills, B. A., "Decompression Sickness: A Study of Cavita-
tion at the Liquid-liquid Interface," Aerospace Medi-
cine , pp. 814 to 817, August 1967.
Marsh, M, C., and F. P. Gorhnm, "The Gas Disease in Fishes,"
Report of the Bureau of Fisheries, pp. 343 to 376, 1904.
Meekin, Tom K#J "Levels of Nitrogen Supersaturation at
Chief Joseph Dnm Under Various Spill Conditions, Phase I,"
Washington Dept. of Fisheries Report to Corps of Engi-
neers, Contract No. NPSSU-71-796, 24 pp., April 1971.
Meekir., Thomas K., Richfird L. Allen, and Austin C, Moser,
1lAn Evaluation of the Rocky Reach Chinook Salmon
Spawning Channel, 1961-1968." Washington Depart, of
Fisheries Tec. Rept. of Fisheries Tec. Rept. No. 6.
February 1971.
Pauley, Gilbert B., and Roy E. Nakatani, "Histopathology of
'Gas Bubble' Disease in Salmon Fingerlings," Jour. Fish.
Res. Bd. of Can.. Voa 24. No, 4, pp. 867-871, 1967.
Raymond, H^ L„, "Branded Juveniles Indicate Dams Destruc-
tive," Fish. Business Weekly, January 19, 1970A.
Raymond, H0 L., "A Summary of the 1969 and 1970 Outmigra-
tion of Juvenile Chinook Salmon and Steelhead Trout
from the Snake River," Bur. of Commercial Fisheries
Biological Lab., Progress Report, Seattle, Wash.,
11 pp., September 1970B.
Rucker, R. R., and K„ Hodgeboom, "Observations on Gas Bubble
Disease of Fish," Progressive Fish Culturist. 15:24-26,
1953.
-------
104
Rucker, R„ R„ and E. M0 Tuttle, "Removal of Excess Nitrogen
in a Hatchery Water Supply," Proft. Fish. Cult., 10:88-90
1948.
Scholander, P, F,, L. VanDam, C, Lloyd Claff, and J„ W.
Kanwisher, "Micro Gasoraetric Determination of Dis-
solved Oxygen and Nitrogen," Biol. Bull. 109:328 to
334, 1955.
Schoning, Robert N., "Nitrogen and Fish," A presentation of
the Columbia Basin state and federal fisheries agencies
to the Tri-State Governors Meeting on the Nitrogen
Supersaturation Problem on the Columbia and Snake
Rivers, Portland, Oregon, March 23, 1971.
Shirahata, Soichiro, "Experiments on Nitrogen Gas Disease
with Rainbow Trout Fry," Bull. Fresh. Fish. Res. Lab. 15
No, 2, March 19660
Spink, Thomas Joseph, "The Pressure and Temperature Effects
on the Solubility of Nitrogen in Distilled Water,"
Master's Thesis, Oregon State University, Corvallis,
Oregon, June 1971.
Swinnerton, John W„, Victor J. Linnenbom, and Conrad H.
Cheek, "Determination of Dissolved Gases in Aqueous
Solutions by Gas Chromatography," Anal. Chem,, 34:483
to 485, 1962,
Weiss, R0 F., "Dissolved Gases and Total Inorganic Carbon
in Seawater: Distribution, Solubilities, and Shipboard
Gas Chromatogiaphy," Ph.D„ Thesis, Univ. of Calif.,
San Diego, 1970A.
Weiss, R. F., "The Solubility of Nitrogen, Oxygen and Argon
in Water and Seawater," Deep Sea Research. Vo„ 17, pp.
to 735, 1970B.
Westgard, R. L., "Physical and Biological Aspects of
Gas-Bubble Disease in Impounded Adult Chinook Salmon
at McNary Spawning Channel," Trans. Am. Fish, Soc. 93:
306-309, 1964.
Wyatt, Ellis J., "Studies of the South Santiam Water Supply
System," Fish Commission of Oregon, memo report,
December 1969.
-------
105
glossary y
AVERAGE MEGAWATT - A unit of average energy output over a specified
time period (total energy m megawatt-hours divided by the
number of hours in the time period).
BASE LOAD - See Load, Base.
BRITISH THERMAL UNIT (Btu) - The standard unit for measurement of
the amount of heat energy, such as the heat content of fuel.
Equal to the amount of heat energy necessary to raise the
temperature of one pound of water one degree Fahrenheit.
CAPABILITY - The maximum load which a generator, turbine, power
plant, transmission circuit, or power system can supply
under specified conditions for a given time interval without
exceeding approved limits of temperature and stress.
Maximum Plant Capability (Hydro) The maximum load which a
hydroelectric plant can supply under optimum head and
flow conditions without exceeding approved limits
of temperature and stress. This may be less than the
overload rating of the generators due to encroachment
of tailwater on head at high discharges.
Peaking Capability - The maximum peak load that can be
supplied by a generating unit, station, or system in a
stated time period. For a hydro project the peaking
capability would be equal to the maximum plant capa-
bility only under favorable pool and flow conditions,
often the peaking capability may be less due to reser-
voir drawdown or tailwater encroachment.
Ultimate Plant Capability (Hydro) - The maximum plant capa-
bility of a hydroelectric plant when all contemplated
generating units have been installed.
1/ Reference: Anonymous, 1970
-------
106
CAPACITY - The load for which a generator, transmission circuit,
power plant, or system is rated. Capacity is also used
synonymously with capability.
Dependable Capacity - The load-carrying ability of a station
or system under adverse conditions for the time interval
and period specified when related to the characteristics
of the load to be supplied. Fot hydro projects the
term refers to the capability m the most adverse month
in the critical period--January 1932 in the case of the
1928-32 critical period.
Firm Capacity - Capacity which has assured availability to
the customer on a demand basis. System firm capacity
consists essentially of hydro system dependable capacity
plus thermal plant installed capacity plus firm imports
minus maintenance and forced outage reserves.'
Hydraulic Capacity - The maximum flow which a hydroelectric
plant can utilize for power generation.
Installed Capacity - Same as nameplate capacity unless
otherwise specified.
Nameplate Capacity - The nominal rated capacity of a
generating unil oi otiici biwildi apparatus. The terni
gives an indication of the approximate generating capa-
bility of the unit, but in many cases the unit is capable
of generating on a continuous basis substantially more
than the nameplate capacity (see Overload Capacity,
below).
Overload Capacity - The maximum load that a machine,
apparatus, or device can carry for a specified period
of time under specified conditions when operating beyond
its nameplate rating but within the limits of the manu-
facturer's guarantee or, in the case of expiration of
the guarantee, within safe limits as determined by the
owner. For example, most of the generators installed
in the region's newer hydroelectric plants have a
continuous overload capacity of 115 percent of the
nameplate capacity.
Peaking Capacity - Same as Peaking Capability.
Reserve Capacity - Extra generating capacity available to
meet unanticipated demands for power or to generate
power in the event of loss of generation resulting from
scheduled or unscheduled outages of regularly used
generating capacity.
-------
107
CAPACITY FACTOR - The ratio of the average load on the generating
plant ior the period of time considered to the capacity
rating of the pl.int. Unless otherwise identified, capacity
factor is computed on an annual basis. In this Appendix,
the capacity factor of a hydro plant is based on maximum
plant capability and assumed load equal to the average
annual energy.
CONVENTIONAL HYDROELECTRIC PLANT - A hydroelectric power plant
ivhTdTljnTTEcFl^reamTior only once as i t passes downstream,
as opposed to a pumped-storage plant which recirculates all
or a portion of the streamflow in the production of power.
COORDINATED COLUMBIA RI\ER SYSTEM - Contractually, the system of
hydroelectric projects located on the Columbia River and
major tributaries which arc operated together on a coordinated
basis under the terms of the Pacific Northwest Coordination
Agreement. The term is sometimes used in a more general
sense to include also those projects which are operated by
utilities not participating in the Coordination Agreement.
COORDINATION* AGREEMENT - See pages 31-34.
CRITICAL PERIOD - Period when the limitations of hydroelectric
power supply due to water conditions are most critical with
respect to system energy requirements. For a discussion of
the critical period as it applies to the regional hydro-
electric systemj refer to page 135.
CRITICAL WATER YEAR - A term sometimes used interchangeably with
Critical Period when the critical period falls within one
operating year, "he term will lose all significance when
the system moves into a multi-year critical period (see
page 136).
DEMAND - The rate at which electric energy is delivered to or by a
system at a given instant or averaged over any designated
period of time, expressed in kilowatts or other suitable
units.
DRAWDOWN - The distance that the water surface of a reservoir is
lowered from a given elevation as the result of the with-
drawal of water. In specific cases in this Appendix, draw-
down may refer to the maximum drawdown for power operation,
from normal full pool to minimum power pool. Sometimes
drawdown is also expressed in terms of acre-feet of storage
withdrawn.
-------
108
ELECTRO-PROCESS INDUSTRY - An industry which requires very large
amounts of electricity in manufacturing for heat or chemical
processes fas distinguished from wheel-tuiriiiig ur iiieciidiiicai
applications). Examples are electric furnace steel, aluminum,
and chlorine.
ENERGY - That which does or is capable of doing work. It is measured
in terms of the work it is capable of doing; electric energy
is commonly measured in kilowatt-hours or average megawatts.
Average Annual Energy - Average annual energy generated by
a hydroelectric project or system over a specified
period. In the Pacific Northwest the average output of
most projects is based on the historical flows experi-
enced during the period 192S-58, as modified by appro-
priate irrigation depletions.
Firm Energy - Electric energy which is considered to have
assured availability to the customer to meet all or any
agreed upon portion of his load requirements. Firm
energy is based on certain specified probability
considerations, which, in the Pacific Northwest, are
related to the 1928-58 sequence of historical strcamflows
adopted for making system power regulations. System
firm energy capability includes hydro system prime
energy, thermal plant energy capabilities, and firm
imports.
Prime Energy - Hydroelectric energy which is assumed to be
available 100 percent of the time: specifically, the
average energy generated during the critical period.
Secondary Energy - All hydroelectric energy other than
prime energy: specifically, the difference between
average annual energy and prime energy.
Usable Energy - All hydroelectric energy which can be used
in meeting system firm and secondary loads. In the
early years of this study, it is possible that there may
not be a market for all of the secondary energy which
could be generated in years of abundant water supply
and some of the water may have to be diverted over
project spillways and the energy wasted.
ENERGY DEMAND - Sec Demand.
FIRM - Assured.
-------
109
FIRM 1,0 M) CARRYING CAPM3ILITY (FI.CC) - The firm load that a system
could carry under coordinated operation under critical period
strcamflow conditions with the use of all reservoir storage
(refer to page 33). More specific terms are Firm Energy
Load Carrying Capability (ITLCC) and firm Peak Load Carrying
Capability (FPLCC). The general term refers collectively to
both.
FOREBAY - The impoundment immediately above a dam or hydroelectric
plant intake structure.
GENERATION - The act or process of producing electric energy from
other forms of energy; also the amount of electric energy
so produced.
GIGAWATT - One million kilowatts> or one billion watts.
HEAD
Gross Head - The difference of elevations between water
surfaces of the forcbay and tailrace under specified
conditions. In this Appendix, gross head generally
refers to the difference between normal full pool and
average tailwater.
Net Head (Effective Head) - The gross head less all
hydraulic losses except those chargeable to the turbine.
HEAT RATE - A measure of generating station thermal efficiency,
generally expressed as Btu per (net) kilowatt-hour. It is
computed by dividing the total Btu content of the fuel
burned (or of heat released from a nuclear reactor) by the
resulting net kilouatt-hours generated.
HYDRAULIC CAPACITY - See Capacity, Hydraulic.
INDEPENDENT RESOURCES (HYDROELECTRIC) - The hydroelectric projects
of the region vhirh arp nnt included in the Coordinated
Columbia River System (see page 135).
IMPORTS - Power imported from outside the Columbia-North Pacific
Region system being considered, m this Appendix.
INTERT1E - See Transmission Interconnection.
KILOWATT (kw) - The electrical unit of power which equals 1,000
watts or 1.341 horsepower.
KILOWATT-HOUR (kuh) - The basic unit of electrical energy. It
equals one kilowatt of power applied for one hour.
-------
110
LOAD - The amount of pohci* delivered to a given point.
Base Load - The minimum load in a stated period of time.
Firm Load - That part of the system load which must be met
with firm power.
Peak load - Literally, the maximum load in a stated period
of-time. Sometimes the term is used in a general
sorise to describe that portion of the load above the
base load.
LOAD FACTOR - The ratio of the average load over a designated
period to the peak load occurring in that period. In this
Appendix the term applies to annual load factor unless
otherwise specified.
LOAD SHAPE (1.0\D PVTTPRN") - The characteristic variation in the
magnitude ol the power load with respect to time. This
can be for a daily, weekly, or annual period.
LOSSES (ELECTRIC SYSTEM) - Total electric energy loss in the
electric system. It consists of transmission, transformation,
and distribution losses and unaccounted-for energy losses
between sources of supply and points of delivery.
MEGAWATT (row) - One thousand kilowatts, or one million watts.
MEGAWATT-HOUR (riwh) - One thousand ki lowatt-hours„ or one million watt-
hours.
NORMAL FULL ^00^ - Ths forcbE*r surfsc? 6l?v£tion
within the reservoir's normal operating range.
NORTHWEST POWER POOL - See page 27.
PACIFIC NORTIIKEST COORDINATION AGREEMENT - See pages 31-34.
PEAK LOAD - Sec Load, Peak.
PEAKING - Power plant operation to meet the variable portion of
the daily load. See Load, Peak.
PEAKIN'G PLANT - A power plant which is normally operated to
provide all or most of its generation during maximum load
periods.
PENSTOCK - A conduit to carry water to the turbines of a hydro-
electric plant (usually refers only to conduits which are
under pressure).
PLANT FACTOR - Same as Capacity Factor.
PONDAGE - Reservoir power storage capacity of linv.ted magnitude
that provides only daily or weekly regulatLon of streamflow.
-------
Ill
POWER - The time rate of transferring energy. Note--The term is
frequently used in a broad sense, as a commodity of capacity
and energy, having only general association with classic or
scientific meaning.
Firm Power - Power which is considered to have assured
availability to the customer to meet all or any agreed
upon portion of his load requirements. It is firm
energy supported by sufficient capacity to fit the load
pattern. The availability of firm power is based on
the same probability considerations as is firm energy.
Interruptib1c Power - Nonfirm power; power made available
under agreements which permit curtailment or cessation
of delivery by the supplier. In the Pacific Nortinvest,
interruptible power loads are met with secondary hydro
energy.
Prime Power - Prime energy shaped to fit the regional load
pattern.
Secondary Po^er - Same as Secondary Energy.
POWER DEMAND - See Demand.
PUMPED STORAGE PLANT - A hydroelectric power plant which generates
electric energy for peak load use by utilizing water pumped
into a storage reservoir during off-peak periods. Refer
also to pages 85-87.
REGULATION (Hydroelectric System) - See pages 134-135.
RESERVES
Reserve Generating Capacity - See Capacity, Reserve.
Spinning Reserve - Generating capacity connected to the
bus and ready to take load. It also includes capacity
available in generating units which are operating at
less than their capability.
RUN-OT-RIVER PLANT - A hydroelectric plant which depends chiefly
on the flow of a stream as it occurs for generation, as
opposed to a storage project, which has sufficient storage
capacity to carry water from one season to another. Some
run-of-river projects have a limited storage capacity
(pondage) iihich permits them to regulate streamflow on a
daily or weekly basis.
-------
112
STORAGE
Dead Storage - The voluinc of water remaining in a reservoir
after a] J of the usable storage has been withdrawn,
(int,ii)- flic toi.il vo]imic of wfiler in a reservoir at
ft or1? 11 J i11! 1
Seasons! Storage - li'atcf lie Id aver from the annual higli-
water season to t]ie fallowLng. lou-water season.
Usable Storage - The volume of storage in a reservoir which
can be withdrawn for various conservation purposes (gross
storage minus dead storage). As used in this Appendix
the term refers to storage which can be withdrawn either
jointly or exclusively For power generation-
STORAGE PfKIJrCT - A project wit}) a reservoir of sufficient size to
carryover from the high-flow season to the low-flow season
an4 thus to develop a firm flow substantially more than the
taiftimm watuv^l flow. i\ stcraje prefect its ovjti
power plant or may be used only for increasing generation
at downstream plants.
TAILWATER - The water surface immediately downstream ffom a dam or
hydroelectric powerplant.
THERMAL PI AMT - A po-er generating plant .ahich uses fvejt to produce
energy. Such plants may burn fossil fuels or use nuclear
energy to produce the necessary thar-nal orergy.
TRANSMISSION GRID - An interconnected system of electric transmission
Uries and associated equipment for the movement or transfer
of electric energy m bulk between points of supply and
points of demand.
TRANSMISSION lirnSCDlWErtltK - Transmission cvroiVL xised
to"tie or interconnect two load areas or two utility systems.
ULTIMATE DEVELOPMENT - The maximum contemplated generating
installation at a power plant.
-------
113
The following references were used in the preparation of
the glossary.
1. Columbia-North Pacific Technical Staff, Glossary: Columbia-
North Pacific framework Study (Publication C-NP 11),
December 1967, Vancouver, Washington
2. Federal Power Commission, Glossary of Important Power and Rate
Terms, Abbreviations, and Units of Measurement (Prepared under
the direction of the Inter-Agency Committee on Water Resources
and promulgated by the Federal Power Commission). 1965,
Washington, D. C.
3. Northwest Power Pool, Glossary of Terms Used in Coordinated
Operation, January 1959, Portland, Oregon
4. Pacific Northwest Coordination Agreement, Agreement for
Coordination of Operations Among Power Systems of the
Pacific Northwest, September 1964, Portland, Oregon
The largest share of the definitions were drawn from the Federc.1
Power Commission glossary, but in a number of cases these definitions
had to be modified or supplemented to reflect regional usage of the
terms.
-------
114
1/
APPENDIX ~
Pumped Storage
"Electrical resource studies indicate that, in the future,
a major part of the Pacific Northwest's base load will be met
by nuclear powerplants. Nuclear plants can supply base load
energy at a relatively low cost, but are an expensive source
of peaking power. Therefore, more economical means for pro-
viding peaking power must be sought. Studies indicate that
the peaking requirements of the region will be met until
about 1990, by adding generating units at existing conventional
hydroelectric projects. When addition of those units is com-
pleted, other sources of peaking power must be developed.
Several alternative sources are available, including pumped-
storage. Recent improvements in reversible pump-turbines
have created considerable interest in pumped storage, especially
in areas where reservoir sites with high head are available,
as they are in the Columbia-North Pacific Region.
Operation
"Pumped-storage hydro is unique among methods of power
generation as it is dependent on other electrical power sources
for its energy supply. It functions as an energy accumulator
in that low-valued off-peak energy is stored by pumping water
from a lower to a higher reservoir (figure 7) [figure 7 not in
this report] . The stored water can then be returned through the
turbines to generate power during peak load periods, when it
is most needed and has its greatest value. Pumped-storage
installations offer many of the favorable characteristics of
conventional hydroelectric plants including rapid start-up,
long life, dependability, low operating and maintenance costs,
and adaptability as low cost spinning reserve. Due to trans-
mission losses and inefficiencies in the operation of pumped-
turbines, approximately one and one-half times as much energy
is required for pumping as is obtained in the generating phase.
However, this increased energy use is justified by the high
value of the peak generation.
JL/ Quoted from (Anonymous, 1970)
-------
115
"Pumped-storage may be designed to operaf"' on a seasonal,
weekly, or a daily cycle. Seasonal pumped-storage would be
economical only in a system where there is a period in the
year in which there is both surplus water and surplus energy.
The surplus energy would be used to pump the surplus water into
a holding reservoir to be used for generation during periods
of greatest power demand. Projects of this type are especially
adaptable to multi-purpose development. An example of this is
the Paterson Ridge site, which is being considered as a
multiple-purpose project for irrigation, recreation, thermal
plant siting, and seasonal pumped-storage power generation.
"Daily and weekly pumped-storage hold considerable promise,
especially in light of the fact that in the near future thermal
plants will begin assuming an increasing share of the region's
base load. As more thermal plants are put into operation,
more off-peak energy will become available for potential use
by pumped-storage plants. Water would generally be pumped at
night (and on weekends) and released during the day to generate
energy for meeting the system's peak loads.
"Pumped-storage projects are generally classified as either
"pure" pumped-storage or "combined" pumped-storage, A "pure"
pumped-storage project is one which operates exclusively as
a pumped-storage plant. The plant's generation capability is
dependent wholly on water pumped from the lower to the upper
reservoir. On the other hand, a "combined" pumped-storage
project is a conventional hydro project whose generating plant
consists either partially or wholly of reversible pump-turbines.
Water pumped from the lower pool serves only to supplement
conventional reservoir inflow as a source of energy.
Summary
"It appears . . . that there is considerable pumped storage
potential in the Columbia-North Pacific Region. In western
Oregon and Washington alone there are nearly 300 sites worthy
of consideration for development as 1,000 megawatt daily/weekly
cycle peaking plants. Some of these sites possibly could be
developed up to 10,000 megawatts. . . . indications are that
there are also a few seasonal pumped-storage sites available
and a number of convential hydro projects (existing as well as
proposed) which could be developed as combined pumped-storage
plants.
-------
116
"At the pre.scnt.ly projected rate of power development,
supplemental peaking capacity beyond that available for
development at conventional hydro projects will probably not
be required until after 1990. However, as thermal plants
are developed, the inexpensive off-peak energy which will
be available may encourage the early development of some
pumped-storage before 1990o An indication of the current
interest in this is the fact that reversible units are now
being installed at an existing project and that detailed
studies are now underway on at least five additional pumped-
storage sites. Furthermore, studies may show that pumped-
storage would be more economical than some of the conventional
hydro units currently scheduled to carry the additional peaking
loads developing prior to 1990o For example, it may be found
that it would be more economical to utilize the limited storage
in The Dalles pool for pumped-storage operation rather than
for handling the releases from the proposed John Day units
#17-20,
"Studies are now being conducted to determine when and
how pumped storage will best fit into the regional load curve.
These studies should also give an indication as to the type
and number of plants which will be required by the years 2000
and 2020. More studies will be required to determine the
relative desirability of individual sites and the effects
that the operation of these plants will have on their environ-
ment, but it is evident that pumped-storage offers considerable
promise as a source of future peaking capacity."
------- |