Ecological Research Series
SALMONID BIOASSAY OF
SUPERSATURATED DISSOLVED AIR IN WATER
Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Duluth. Minnesota 55804
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and avmaximum interface in related fields.
The five series are:
1.. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal
species, and materials. Problems are assessed for their long- and short-term
influences. Investigations include formation, transport, and pathway studies to
determine the fate of pollutants and their effects. This work provides the technical
basis for setting standards to minimize undesirable changes in living organisms
in the aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/3-76-056
July 1976
SALMONID BIOASSAY OF SUPERSATURATED
DISSOLVED AIR IN WATER
by
Earl Dawley
Bruce Monk
Michael Schiewe
Frank Ossiander
W. Ebel
National Marine Fisheries Service
Northwest Fisheries Center
Seattle, Washington 98112
EPA-IAG-0155(D)
Project Officer
Gerald R. Bouck
Western Fish Toxicology Station*
Corvallis, Oregon 97330
(*Foraerly with the Environmental Research Laboratory -
Duluth, now with the Corvallis Environmental Research
Laboratory, Corvallis, Oregon 97330)
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
ENVIRONMENTAL RESEARCH LABORATORY
DULUTH, MINNESOTA 55804
-------
DISCLAIMER
This report has been reviewed by the Environmental Research Laboratory -
Duluth, U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views
and policies of the U.S. Environmental Protection Agency, nor does mention
of trade names or commercial products constitute endorsement or recommenda-
tion for use.
ii
-------
ABSTRACT
Tests were conducted in shallow (.25 m) and deep (2.5 m) tanks of water
at 10°G with concentrations of dissolved atmospheric gas ranging from
100 to 127% of saturation to determine the lethal and sublethal affects
of the dissolved gas on juvenile fall chinook salmon, Oncorhynchus
tshawytscha, and steelhead trout, Salmo gairdneri.
Fall chinook salmon (average fork length of 42 mm) were much more re-
sistant to supersaturation than juvenile steelhead trout (average fork
length of 180 mm). Salmon tested in the shallow tanks at 120% of
saturation incurred 50% mortality after 22 days whereas trout tested
at the same level, incurred 50% mortality in 30 hrs. Signs of gas bubble
disease were noted on dead fish and on subsamples of live fish from
deep water tests at 1107o saturation and above and in shallow water tests
at 1057o and above. Vertical distribution of both salmon and trout in
the deep tanks appeared to compensate for about 107<, and 10 to 1570 re-
spectively of effective saturation. Average depths of the fish in deep
tanks increased with increased gas concentration. Significant differ-
ences in growth and condition factor of the salmon and trout were not
found between stressed and control fish during the test period.
iii
-------
CONTENTS
Page
Abstract ill
List of Figures vi
List of Tables viii
Acknowledgments ix
Sections
I Conclusions 1
II Recommendations 2
III Introduction 3
IV Methods and Materials 4
V Results 13
VI Discussion 30
VII Literature Cited 36
V
-------
FIGURES
No. Page
1 Schematic drawing of system used to produce water 8
supersaturated with dissolved atmospheric gas in
shallow water tanks.
2 Mortality versus time curves for juvenile fall 14
chinook salmon exposed to various concentrations
of dissolved atmospheric gas (% of saturation T.D.G.)
in shallow (0.25 m) and deep water (2.5 m) tanks at
10°C.
3 Mortality versus time curves for juvenile steelhead 15
trout exposed for 7 days to various concentrations
of dissolved atmospheric gas (% of saturation T.D.G.)
in shallow (0.25 m) and deep (2.5 m) water tanks at
10°C.
4 Frequency (%) of dead juvenile chinook salmon with 17
signs of gas-bubble disease for shallow (0.25 m)
test tanks with different concentrations of dissolved
gas.
5 Exposure times of 25% mortality of juvenile chinook 19
salmon held at different concentrations of dissolved
gas in deep (2.5 m) and shallow test (0.25 m) tanks.
6 Frequency (%) of dead juvenile chinook salmon with 20
selected signs of gas-bubble disease for shallow
(0.25 m) and deep (2.5 m) test tanks with different
concentrations of dissolved gas.
7 Mean depth during daylight hours of groups of juve- 21
nile chinook salmon held in 2.5 m deep tanks at
dissolved gas concentrations of 100? 105f 110, 115f
120^ 124f and 1277» of saturation—averaged for per-
iods of 10 days over 127 days.
vi
-------
No. Page
8 Mean depths of groups of juvenile chinook salmon 22
in 2.5 m deep tanks at dissolved gas concentrations
of 100, 105r 110, 120, 124, and 127% of saturation-
averaged for 3 periods of the day over 30 days.
9 Mean depth of groups of juvenile steelhead trout 23
in 2.5 m deep tanks at dissolved gas concentrations
of 110, 115, 120, and 127% of saturation—averaged
for 2 periods of the day over 15 days.
10 Mortality versus time curves for fall chinook salmon 33
at various size (40-67 mm forklength) when exposed
to dissolved gas at 110, 112, and 115% of saturation
T.D.G. in shallow tanks (0.25 m or less). Tests at
112% T.D.G. were conducted by Meekin and Turner
(1974).
vii
-------
TABLES
No. Page
1 Length, weight, and condition factor of randomly 5
sampled fall chinook salmon and steelhead trout
taken from test populations before testing.
2 Chemical features of the water in the test tanks, 10
20 February to 25 June 1973 and 6 to 21 May 1974.
3 Mean concentrations of dissolved gas in test tanks 12
over test periods of 4 mo. for chinook salmon and 7
days for steelhead trout.
4 Estimated values for model parameters, 60-day survival 29
probability and expected life for a 25% survival
proportion.
viii
-------
ACKNOWLEDGMENTS
We extend our sincerest appreciation to the Washington State Depart-
ment of Ecology, especially Stephen Robb for the efforts taken to solve
many water quality problems encountered before these tests and for
the monthly analysis of our water system.
Also, thanks to Dr. Gerald Bouck for his tolerance with the problems
that arose during the study and for concern and help given during
organization and implementation.
ix
-------
SECTION I
CONCLUSIONS
(1) At 115% of saturation in shallow tanks (where hydrostatic compen-
sation is not possible) substantial ( 807°) mortality to fall chinook
populations occurred within 60 days of exposure. After 60 days of
exposure, significant (13%) mortality occurred among salmon in the
shallow tanks at 110% of saturation whereas in the deep tanks at 120%
of saturation 4% mortality occurred.
(2) Substantial (757>) mortality of steelhead trout tested in the shallow
tanks occurred at about 1157o of saturation and in deep tanks at about
127% of saturation, 25% mortality occurred after 7 days exposure.
(3) Tolerance to supersaturation of atmospheric gas decreased with
age and growth in both species.
(4) Emboli in branchial arteries, gill filaments, and the heart
were rarely observed on live subsamples but were prevalent on dead
fish, indicating these signs are directly associated with the death
of the animal.
(5) The average depth maintained by salmon and trout when allowed
to sound compensated for about 10% and 10 to 15% (respectively) of
the saturation value measured and computed on the basis of surface
(760 mm Hg) pressure.
(6) Salmon and trout with signs of gas-bubble disease were able to
recover from exposure to supersaturation.
(7) Exposure to various concentrations of supersaturated gas does
I
not seem to affect the ability of steelhead trout to adjust to salt
water; data on the ability of fall chinook to adjust to salt water
were inconclusive.
-------
SECTION II
RECOMMENDATIONS
On the basis of the information contained in this report we recommend:
(1) that the provisional dissolved atmospheric gas standards remain
as it is (110%) unless substantial depth distribution information on
wild stocks of fish indicates that the level can be safely raised.
(2) that the dissolved atmospheric gas level be held below 110% of
saturation whenever fish are confined in shallow water less than .25 m
deep.
(3) that additional studies be done to determine whether the tendency
of chinook and steelhead trout to sound when exposed to supersaturation
occurs in natural waters.
-------
SECTION III
INTRODUCTION
Effects of supersaturation of dissolved atmospheric gases on fresh
water fishes have been studied by many investigators since the late
1800's. The current problem of supersaturation in the Columbia River
and its large tributary the Snake River (Ebel, 1969; Beiningen and
Ebel, 1970; Ebel, 1971; Meekin and Allen, 1974) has resulted in
renewed interest in effects of supersaturation on fish, and a great
deal of research on the subject has recently been accomplished by
fishery and other agencies in the Pacific Northwest. Information
on the resistance of indigenous fish species to high gas concentra-
tions is we11-documented for exposure in shallow water for short
periods of time (Rucker and Tuttle, 1948; Harvey and Cooper, 1962;
Coutant and Genoway, 1968; Bouck et al., 1970; Ebel, Dawley, and
Monk, 1971; Bouck, 1972; Blahm, McConnell, and Snyder, 1973; Fickei-
sen, Montgomery and Schneider 1973; Dawley and Ebel, in press) but
there are still many unanswered questions regarding the effects
of chronic low level exposure on survival. Fish may be subjected
to low levels of supersaturation in two ways: they may inhabit water
areas where they cannot compensate for gas saturation by sounding
or they may inhabit deep water areas where hydrostatic pressure
offsets the effects of high gas levels.
r
The National Marine Fisheries Service, funded in part by the Environ-
mental Protection Agency (EPA) in 1972, began investigations of
chronic effects of long term exposure of juvenile fall chinook to
various low levels of supersaturation. In this report we describe
those effects observed from deep and shallow water tanks on juvenile
fall chinook salmon Oncorhynchus tshawytscha, and juvenile steelhead
trout, Salmo ^ardneri.
-------
SECTION IV
METHODS AND MATERIALS
Chinook salmon and steelhead trout are valuable species in the Columbia
River system, and particularly in the Snake River and were used
in the following study. Juvenile fall chinook were acquired for
our experiment as buttoned up fry from the Spring Creek National
Fish Hatchery in early February 1973. Juvenile steelhead were captured
during their seaward migration down the Snake River on April 30,
1974. Steelhead were one plus years of age. Chinook and steelhead
populations were acclimated to our laboratory water system at 10°C,
for 19 and 6 days respectively prior to testing. Before beginning
the tests, random samples were taken from each population to obtain
average weights, lengths and condition factors (Table 1). Fish were
fed an Oregon Moist Pellet ration 5 days a week, at a rate of 4%
of body weight/day. Rations for each test tank were corrected daily
for numbers of surviving fish and corrected each 28 days for weight
change (calculated from size of fish subsampled every four weeks).
Two series of tests on effects of dissolved gas were conducted at
the Northwest Fisheries Center (Seattle, Washington). The first
was completed in 1973 using the fall chinook salmon as experimental
animals, and the second in 1974 using the steelhead trout. The series
consisted of 20 simultaneous tests of chinook and 18 simultaneous
tests of steelhead, in fresh water at 10°C, with various concentrations
of dissolved gas in deep and shallow water tanks. At termination
of the tests, surviving fish were divided into two groups; one group
was transferred directly to salt water to assess the effects of
stress from supersaturation on their ability to acclimate to salt
water; a second group was examined for signs of gas-bubble disease.
Groups of chinook salmon and 1 group of steelhead exhibiting signs
were then placed in equilibrated water (10070 of saturation total
dissolved atmospheric gas) for a 2 week recovery period and subse-
quently re-examined for signs.
-------
Table l.—LENGTH, WEIGHT, AND CONDITION FACTOR OF RANDOMLY SAMPLED FALL CHINOOK
SALMON AND STEELKEAD TROUT TAKEN FROM TEST POPULATIONS BEFORE TESTING.
FALL CHINOOK
Ho. of
Species fish
Chinook salmon 60
81*
Steelhead trout 2h
26
29
28
I/ Condition factor •
Ave
of
.1*3
.te
56.
5*.
53.
5U.
v (weight
. weight
fish
5
5
7
8
.06
.06
STEELHEAD
13.3
13.7
1U.6
Ave. length
of fish
(mm) s.d.
UO
UO
180
180
180
180
1.3
1.2
1U
15
15
16
Ave.
cond.
.67
.65
.922
.890
.907
.917
I/
factor
in grams) x 10^
L-^fork length in millimeters^
-------
Water in the deep tanks was 2.44 m (8 feet) which provided a maximum
hydrostatic compensation from excess dissolved gas of 0.24 atm.or
24% of saturation total dissolved gas (T.D.G.); and in shallow tanks
0.25 m (10 inches) deep providing only 0.025 atm. of pressure compensa-
tion or 2.5% of saturation. The shallow water tests on both chinook
salmon and steelhead trout consisted of two replicates at 120, 115,
110, 105 and 100% (control) of saturation T.D.G. Deep water tests
with chinook salmon included tests at 127% (1 tank), 124% (1 tank),
120% (2 tanks), 115% (2 tanks), 110% (2 tanks), 105% (1 tank) and
100% (1 tank). Deep water tests with steelhead trout consisted of
two replicates at 127%, 120%, 115% and at 110% T.D.G. (Previous
studies indicated that tests at 110% of saturation in deep tanks
could serve as a quasi-control.
The tests with chinook salmon began February 20, 1973 with the intro-
duction of 220 fish per tank and was terminated on July 8, (127
days of exposure to concentrations of excess dissolved gas plus
13 days of subsequent tests). Steelhead tests began May 6, 1974
with the introduction of about 80 fish per tank; these were terminated
after 21 days (7 days of exposure to supersaturation of dissolved
gas plus 14 days of subsequent tests).
Once testing began, each tank was examined 4 times daily for the
first 4 days followed by three, two, or one times daily throughout
the remainder of the test period. During each observation dead fish
were removed, their length and weight recorded, and signs of gas-bubble
disease noted. Vertical distribution of the fish in each deep tank
was also noted in percentage of total population at four levels
of depth; 0-0.6, i6-1.2 m, 1.2-1.8 m, and 1.8-2.5 m. Subsampling
of each test group for condition factor and disease signs was done
each 28 days at a rate of 10% (but not less than 5 individuals)
of the surviving population. Fish from each subsample were weighed,
measured, and examined for signs of gas-bubble disease (none were re-
turned to the tests).
-------
DechLorinated water from the public water system of Seattle--which
came from the Cedar and Tolt Rivers—was used in these tests. Tempera-
ture was maintained at 10° + ,5°C, by mixing hot (27°C) and cold
(7°C) water in a reservoir tank. Water for the shallow tank system
was supersaturated by injecting 0.5 1/min air and 0.23 1/min 0-
into the suction side of two centrifugal pumps which were plumbed
with a recirculation loop to 2 closed receivers [197 liters (52
gal) each]. Hydraulic pressure within the receivers was maintained
2
at 2.1 kg/cm (30 psi) where dissolved gas content was increased
to about 122% of saturation T.D.G. (Fig. 1). Water for the deep
tank system was recirculated by a pump through an open reservoir
tank 9 m deep x 3 m in diameter which was tapped at the bottom for
distribution to the test tanks. Air and oxygen were injected into
the recirculating pump at about 2.0 1/min and 0.2 1/min respectively.
This resulted in a stable saturation level of 128% T.D.G. Both sources
supplied individual test tanks through polyvinyl chloride (PVC) pipes
which directed the supersaturated water to a vertical stack of alun-
inum trays (each 28 x 41 cm) placed 5 to 10 cm above one another^ One
half of each tray was perforated with 500 - 3 mm holes and the per-
forated ends alternated to produce a back and forth flow of water
from tray to tray. Water was then collected in a plexiglass box (46 x
46 x 25 cm deep) with a false bottom of porous polyethylene plate
through which air was passed. The level of supersaturation desired for
each test tank was maintained by regulating the number of perforated
trays and the amount of air supplied to the collection boxes (greater
contact with air at atmospheric pressure decreased the level of
supersaturation). Water from each box was gravity fed to a test
tank through a vinyl tube, ending at the water surface. A flow rate
of 7.5 1/min was maintained which created a circulation of about
0.2 m/sec within the test tank.
Test tanks were made of green tinted fiberglass, 1.2 m in diameter
and of two heights, 0.6 m and 3.0 m (shallow, deep) holding about
270 and 2700 liters of water, respectively. A plexiglass window
-------
Bock pressure
volve
00
Equilibration
troys
Air equilibrotor
Centrifugal pump
PRESSURE
TANK
Recirculate
SHALLOW TANK
Fig. 1—Schematic drawing of system used to produce water supersaturated with dissolved
atmospheric gas in shallow water tanks.
-------
extended from the top to the bottom of the deep tanks allowing obser-
vations to be made over the entire water column. Curtains covered
the windows and were removed only at times of observation. Water
drained from the bottom of these tanks through an external stand
pipe.
Lighting was controlled with time clocks to simulate natural sunrise
and sunset and light intensity at the surface of each test tank
was from 108 to 216 lumen/sq meter during full intensity periods.
Data on water quality (other than information on dissolved oxygen
and nitrogen plus argon) were obtained before testing began and
once each week—or once every 4 weeks depending on the chemical
feature measured (Table 2). Procedures followed during analysis
of water were those of the American Public Health Association, et
al.j(1971). The monthly measurements were made by personnel of the
State of Washington Department of Ecology using a Perkin - Elmer
303 atomic absorption spectraphotometer.* All concentrations of
heavy metals or other potentially dangerous compounds fell far below
danger levels to chinook salmon and steelhead trout (McKee and Wolf,
1963).
Dissolved gas analyses were made on water from each tank at least
once each weekday for the first two weeks, then a minimum of twice
each week for the rest of the test period. Procedures were identical
to Dawley and Ebel (in press); they employed gas chromatography
calibrated for nitrogen plus argon using a modified manometric blood
gas analysis apparatus (Van Slyke and Neill, 1924). The modified
Winkler procedure (American Public Health Association e,t al., 1971)
was used for analysis of oxygen concentrations. Water samples were
collected by use of a siphon tube from the middle of the water column
*Trade names referred to in this publication are not an endorsement of
commercial products by the Natl. Mar. Fish. Serv.
-------
Table 2.—CHEMICAL FEATURES (COUCEHTRATIOHS IN mg/L) OF THE MATER III THE TEST TAJIKS, 20 FEBRUARY TO
V
25 JUIIE 1973 and 6 T.O 21 MAY 1971*.
CHEMICAL Shallov tanks (gag concentration) ~ Deep tanks (gas concentration)
FEATURE IQOg ..105?, 110* 115i* __120X' _ IPOS 105JJ 110? __' 113* 1203 125? 128?
Weekly Measurements
Tot. Hard. 18-21 20 19-21
Tot. Alk. 10-15 32-13 10-13 >
pH 7.0-7.3 6.9-7.1 6.8-7.1 6.8-6.9 6.9-7.1 6.9-7.3 7.0r7.3 6.9-7.1 6.8-7.1 6.7-6.9 6.7-7.1 6.7-7.
<.05^ <.05 <.05 <.05 <.05 <.05 <.05 <.05 <-05 <.05 <.05 <.05
<.02 <.02 <.02
Monthly Measurements
I/
Zn n.d. n.d.
Cu n.d. n.d.
Cd n.d. n.dr
Pb n.d. n.d.
CQ 3.3-U.4 3.3-U.-3
I/ Other parameters measured prior to testing: C02 (1.2-2.0), chloride (U-6). cyanide (<.02), floride (.9-1),
iron (.1-.8), nitrate (.03), nitrite (.003), phenol (.01), potassium (.2-1), sulfate ^3).
Measurements made by Environmental Protection Agencj--, Redmond, Wash.
2/ <—below minimum detectable levels indicated using colorimetric techniques.
3/ n.d.-nondetectable on I^rkin-ELner 303 atomic absorption spectrophotometer Detection limits were
Zn <0.005, Cu <0.01, Cd<0.025, and Pb ^0.05. Measurements made by Wash. State Dept. of Ecology.
-------
in the shallow tanks and from the surface of the deep tanks (clear
vinyl tubes were used because some latex tubing produced a degassing
effect as the water was siphoned). Gas concentrations remained steady
throughout the test periods and mean values for each tank did not
change more than 17» on a weekly basis with standard deviations for
both tests less than 2,67o T.D.G. over all (Table 3). Samples were
taken from the topf middle and bottom of the water column of the
deep tanks several times and gas concentrations were found to be
uniform throughout the tank.
11
-------
TABLE 3.—MEAN CONCENTRAT1ON OF DISSOLVED GAS IN TEST TANKS OVER TEST PERIODS OF 4 MONTHS
FOR CHINOOK SALMON AND 7 DAYS FOR STEELHEAD TROUT.
Type
of
tank
Shallow
Deep
Chinook tests
(mean *L of saturation)
Test
no.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
It
18
19
20
°2
122.5
120.9
115.3
115.6
108.3
107.7
102.2
102.5
98.0
98.6
101.7
101.7
108.1
107.7
115.3
114.6
118.6
118.7
124.8
126.8
N2 + Ar
120.0
120.0
115.6
115.9
110.9
109.9
104.9
105.3
100.4
100.5
98.6
105.3
111.2
110.8
116.0
115.9
119.7
119.6
124.4
127.1
T.D.G. -
120.4
119.9
115.2
115.4
109.8
109.3
104.1
104.4
99.8
100.0
100.8
104.2
110.3
110.0
115.6
115.4
119.2
119.2
124.1
126.8
sd
1.2
1.3
1.1
1.4
1.2
1.1
1.3
1.0
1.1
1.0
1.1
1.2
0.8
0.9
1.1
1.0
1.2
1.1
1.6
1.8
Steelhead tests
(mean 2 of saturation)
°2
134.6
132.9
121.1
119.3
108.6
108.1
104.0
100.8
88.3
86.0
-
-
105.6
108.5
114.0
116.3
126.7
125.5
133.4
133.0
N2 + Ar
120.0
119.1
114.4
114.1
109.6
110.2
107.2
107.1
99.8
99.7
-
-
110.2
111.0
114.1
115.0
120.4
119.6
125.2
125.9
T.D.G. —
122.7
121.6
115.6
114.9
109.2
109.5
106.5
105.7
97.3
96.9
-
-
109.1
110.3
113.8
115.0
121.4
120.5
126.6
127.0
- sd
1.5
1.3
1.0
1.4
1.1
1.2
1.6
2.2
1.5
1.0
-
-
1.4
1.0
1.5
0.9
0.8
2.1
2.5
2.0
12
-------
SECTION V
RESULTS
LETHAL EFFECTS QF DISSOLVED GAS
Juvenile chinook salmon held at 120% and 115% of saturation in the
shallow tanks and at 127% and 124% in the deep tanks sustained sub-
stantial mortality (67-97%) after 60 days of exposure. By the same
time, 15% mortality had occurred on fish held at 110% in shallow
tanks and 5% had occurred in those held at 120% in deep tanks (Fig.
2). Mortality was insignificant for salmon held at lower gas concen-
trations. Curves of accumulative mortality of fish in the control
tanks was minimal (3%) for the first 60 days, but by day 127 it
had sharply increased to 26.37» in the shallow tanks and 13.670 in
the deep tanks.
A change in normal feeding response and swimming behavior of chinook
salmon in deep and shallow tanks was noticed on day 64 of the test.
We believe these changes resulted from an infection caused by Cyto-
phaga psychrophila. All fish (test and control) were taken from
tanks and bathed in a 10 ppm solution of terramycin for 1 hour.
A supplement of .5% oxytetracycline was added to the daily ration
of food for the following 10 days, and after a 2 week interval another
.5% supplement was added for 10 days. After the second treatment
(day 100) the fish in all tanks behaved normally.
Steelhead trout held at 120% and 115% of saturation in shallow tanks
developed substantial mortality—100% and 75% respectively—within
7 days and the two groups of fish at 127% in the deep tanks averaged
25% mortality (Fig. 3). Mortality of control fish was 2%, this value
was small in relation to the test groups and was not used to adjust
mortality curves of test groups.
PROGRESSION OF GAS-BUBBLE DISEASE
The frequency of occurrence of most signs of gas-bubble disease
increased with increasing levels of supersaturation. The most fre-
13
-------
120%
Shollow Tonks
Desp Tanks
20
120
KO
40 60 ' 80 ' 100~
DAYS OF EXPOSURE
Fig. 2—Mortality versus time curvesM"or juvenile fall Chinook
salmon exposed to various concentrations of dissolved
of dissolved atmospheric gas (% of saturation T.D.G.)
in shallow (0.25 m) and deep water (2.5 m) tanks at
10° C.
•=> Combined replicates corrected daily for mortality
of control tests by subtracting percentage of
mortality. Cold water disease affected the test
fish on day 6U and ensuing treatment lasted
through day 100.
14
-------
0
0
15%
___0 o 127%
2345
DAYS OF EXPOSURE
105%
15%
10%
Fig. 3—Mortality versus time curves for juvenile steelhead trout exposed
for 7 days to various concentrations of dissolved atmospheric gas
(% of saturation T.D.G.) in shallow (0.25 m) and deep (2.5 m) water
tanks at 10° C.
-------
quently occurring signs of gas-bubble disease were cutaneous gas filled
blisters on the head and mouth and occlusions of gill filaments on salmon
which had succumbed to high gas levels. Dead fish from 120% and
115% shallow, and from 127% deep tanks showed 40-70% incidence of
those signs. Other signs which increased with increasing levels
of supersaturation but with lower frequencies of occurrence were:
heart occlusions (14 to 34%), gas-blisters in the connective tissue
surrounding the eye, and blisters between the fin rays (Fig. 4).
Research by Dawley and Ebel (in press) showed that the appearance
of gas bubbles in the lateral line was the first external sign of
gas-bubble disease to develop on juvenile spring chinook salmon
exposed to high concentrations of dissolved gas. We, however, observed
that gas bubbles in the lateral line of dead fall chinook and steel-
head trout was not prevalent within any group, but did appear in
a high percentage (50-100%) of live fish in subsamples from all
test groups. Scattered bubbles (covering less than 15% of the lateral
line) also appeared on most dead fish from the control groups. We
cannot account for these bubbles on the control fish; and for that
reason, only ascribed this sign to gas-bubble disease when more
than 157o of the lateral line on test fish appeared occluded.
Disease signs on live fish generally appeared in rates and patterns
similar to those recorded on dead fish removed from the same gas
concentrations. However^, emboli in branchial arteries, gill filaments,
and the heart, were rarely observed on live subsamples, but were
prevalent on fish examined immediately after death, indicating these
signs are directly associated with the death of the animal.
The trends in gas-bubble disease signs noted on steelhead trout
differed from those recorded during tests with the fall chinook
salmon. Two differences were:
1. Heart and gill embolism occurred in almost 100% of the dead steel-
head whereas these signs were significantly fewer in dead chinook,
16
-------
80-
2
O
CO
X
i
X
60-
40-
20-
/O
A —A Head
**"" Gills
•O Mouth
/
Abdomen
surface
VLateral line
f?1— 1 —— 1 !——j—
iOO !05 110 115 120
GAS CONCENTRATION (% of saturation )
—i
125
Fig. k—Frequency (%} of dead juvenile chinook salmon with
signs of gas-bubble disease for shallow (0.25 m)
test tanks with different, concentrations of dissolved
-------
suggesting that decomposition of the fall chinook quickly masked
the emboli (due to their small size);
2. Rates of incidence of certain signs were directly associated
with duration of the test. At termination of the tests live and
dead steelhead showed light incidence of exophthalmia, cutaneous gas-
blisters on the head and in the buccal cavityf and no signs of blisters
of the body surface, however, the chinook subjected to supersaturation
for a much longer duration, showed high incidence of those signs.
Also, gas blisters between the fin rays were not prevalent on fall
chinook, yet showed very high incidence on steelhead mortalities
and live subsamples after 7 days exposure to high supersaturation.
EFFECTS OF WATER DEPTH
The average depth of the chinook and steelhead groups in the deep
tanks appears to have compensated for about 10% and 10-15% of the
dissolved gas content respectively; this observation is based on
comparison of rates of mortality between deep and shallow tanks
(Figs. 2 and 3). Figure 5 shows that the time to 25% mortality of
fall chinook at various levels of dissolved gas concentrations in
the deep tanks were comparable to times in the shallow tanks at
a 9.5 to 10% lower effective saturation (i.e., 25% mortality was
reached at 30 days of exposure in the deep tanks at 124%, and in
the shallow tanks at 115%,). Also a comparison of incidence and degree
of gas-bubble disease signs on dead chinook, between deep and shallow
tanks indicates an effective decrease in supersaturation of 12 to
15% of the total dissolved gas concentration (Fig. 6).
Vertical distributions of chinook salmon groups for the first 3
days of testing were variable and not significantly different between
different levels of saturation. After 3 days, however, the fish
held at high saturation test levels maintained a greater average
depth than those at the lower gas levels and maintained this difference
thru the entire test period (Fig. 7). Tests with steelhead trout
showed similar results, i.e., mean depth of fish increased with
18
-------
O
*"•
a
o
vt
127
124
120
DC
S-
UJ
O
8
CO
<
O
!0
105
100
• shallow tanks
odeep tanks
10%
9.5%
\
10%
20 40 60 80 100
DAYS OF EXPOSURE
120
140
160
•Fig. 5—Exposure times of 25$ mortality of juvenile Chinook salmon held at
different concentrations of dissolved gas in deep (2.5 m) and shallow-
test (0.25 m) tanks.
-------
fo
o
CO
z
e>
CO
X
CO
80-i
70
60
50-
40-
30-
20-
10-
0-
100
Deep tanks
Shallow tanks
D Blisters on head
• Blisters in mouth
O Occlusion of gill
filaments
o
105
10
115
120
124
27
GAS CONCENTRATION (% of saturation)
Fig.6--Frequency (%) of dead juvenile Chinook salmon with selected signs of gas-bubble
disease for shallow (0.25 m) and deep (2.5 m) test tanks with different
concentrations of dissolved gas.
-------
.6-
.8-
-S1.CH
o>
o>
x
»-
Q.
CC
UJ
1.8-
110%
19 29
39 49 59 69 79 89
DAYS OF EXPOSURE
99
109
119 ' 129
Fig. 7—Mean depth during daylight hours of groups of juvenile chinook salmon held in 2.5 m deep
tanks at dissolved gas concentrations of 100, 105, 110, 115, 120, 12k, and 127$ of
saturation—averaged for periods of 10 days over 127 days.
-------
NJ
ho
- 0.5
03
QJ
t
bJ
O
or
uu
1.0
2.0
2.5
6
fl
Unlighted A= 1900-0559
bottom
100 105 110 115 120 124 127
GAS CONCENTRATION (% of saturation)
Fig. 8—Mean depths of groups of juvenile Chinook salmon in 2.5 m deep
tanks at dissolved gas concentrations of 100, 105, HO, 120,
12U, and 127$ of saturation—averaged for 3 periods of the day
over 30 days.
-------
t-o
u
0.5
to
-------
increasing gas concentrations. Observations at night on distribution
of fish in the deep tanks showed a downward shift of about 0.3 m
soon after dark (Figures 8 & 9). The increase in average depth of fish
with increasing levels of dissolved gas concentration, which was noted
during daytime observations was also maintained at night.
EFFECT OF GAS SUPERSATURATION ON CONDITION FACTOR
Weight and length data obtained from subsampled fish and freshly dead
test fish were used to calculate a condition factor "K" where:
w (weight in grams) x 10
K = 3
L (fork length in millimeters)
These data were examined to determine if condition factors were related
to susceptibility of fish to gas-bubble disease and death from super-
saturation. Statistical comparisons were made to detect: 1) differences
between condition factors of dead fish and those of live fish within
each test population and; 2) differences among condition factors caused
by chronic exposure to the various levels of supersaturation.
Condition factors and fork length data for salmon that died within 8
days of the monthly subsampling date were compared with those of live
fish within the subsamples. Comparisons were not made if less than five
fish had died within the specified time period. Condition factors
of dead fish from shallow tanks at 110, 115, and 120% saturation and from
deep tanks at 120 124, and 127% saturation were examined (943 fish in
the comparison). Mortalities from 11 test groups were significantly dif-
ferent in condition factor at the 90% probability level (Students T test)
than their live fish counterparts for that test and three groups were not
Significantly different. Means for 12 of the 14 groups were larger for
dead fish than for live fish. Comparisons of fork length measurements for
the same fish groups showed that in 8 of 14 groups dead fish were signifi-
cantly different from the live fish, however, the mean lengths of the
mortalities (12 of 14 groups) were shorter than those of the live fish
groups. Condition factor data may have been extensively biased by weight
change of dead fish from water absorption after death. Some dead fish
-------
remained in the test tanks as long as 16 hrs before being weighed and
measured. Due to the consistency of these length data we believe that
the smaller fish died at a higher rate than the larger fish, and that
condition factor comparisons were invalid. Steelhead comparisons of
lengths, weights, and condition factor showed no difference between
mortalities and survivors of the 7 day test.
A wide range of condition factors of live fish examined each month was
found for groups of chinook salmon tested at the same level of super-
saturation (i.e., between replicates of tests at one gas concentration).
Because of this variability between replicates we could not determine
whether there was a difference in average condition factors (relating
to differences in growth rates) among groups exposed to the various test
levels of supersaturation.
RECOVERY FROM GAS-BUBBLE DISEASE
Upon termination of the tests at various levels of supersaturation,
portions of the groups of chinook salmon and steelhead trout that
exhibited signs of gas-bubble disease were placed in equilibrated
water (100% T.D.G.) tc determine if recovery was possible. Subsamples
of salmon tested at 110% of saturation in shallow tanks at 110,
115, and 120% in deep tanks were used in this phase of the experiment.
A portion of each of these test groups had sustained significant
mortalities from gas-bubble disease. All subsamples (group size
27-48 fish) sustained mortalities from 10~1670 in the two week recovery
period but these could not be attributed to gas bubble disease.
After 2 weeks, the survivors no longer exhibited outward signs with
the exceptions of one fish with a hemmorrhaged eye, and another with
bubbles ir. the orb3t.
Eleven steelhead trout that had survived in the deep tank tests
with a gas concentration of 127% saturation were placed in a shallow
tank at 100% of saturation. After 3 days, the cutaneous blisters
on the trout had decreased in size and number; for example; gas-blisters
5 ram in diameter had decreased in size to 2 mm and 20 blisters on
25
-------
the operculum decreased to 4. These fish were subsequently placed
in water at 105% of saturation and all signs remained the same after
another 4 days, at which time fish were released. There were no
mortalities in the 7 day recovery period.
EFFECTS OF TRANSFER TO SALT WATER
To determine whether prior exposure to various levels of dissolved
gases affected the ability of fish to make the transition to salt
water, subsamples of survivors from combined replicates of all test
groups of chinook salmon and steelhead trout were placed into salt
water at 25 ppt salinity at 10°C. Samples of about 50 chinook salmon
from each saturation level in deep and shallow tanks were transferred
on test day 127 and observed for 13 days. A combined total of 98%
mortality occurred in 3 days. Only 8 salmon survived for the entire
period; 1 fish from the 105% shallow tank and 7 fish from the 110%
deep tank. Results of a statistical comparison of fork lengths of
the survivors (X= 67.3 mm) to those that had succumbed (X= 52. mm)
indicates a definite size correlation with ability to survive transfer
(T= 5.73, 46 df, P 0.01). The majority of the experimental stock
of salmon had not yet reached normal smelting size and their ability
to withstand transfer to saltwater may have been severely lessened.
Subsamples of 10 to 20 steelhead trout from previous tests were
likewise observed after transfer to saltwater. Mortality varied
from 0-16%, but there was no correlation to previous dissolved gas
stress. However, a size comparison between dead and live trout indi-
cated that the dead fish were significantly smaller (T= 1.925, 51
df P<.06) which suggests that the dead fish may not have been up
to smelting size.
MATHEMATICAL MODEL OF MORTALITY
In these tests the fish were subjected to a continuous challenge until
death or the termination of the experiment. The occurrance of a death
was considered an "event" or "point" and the lapses in time between
points constitute a General Point Process. Within this general theory
26
-------
there are several event rate functions which could be considered
candidates to represent the data. The most general event rate function
can accomodate decreasing, constantf and increasing death rates. This
represents the situation where the mortality intensity is changing
with time. In these tests salmon exhibited conditions which could be
"N
characterized by progressively accumulating damage with a consequent
increasing death rate (Figures 2, 4 and 6). If the mortality intensity
is represented as an appropriate power function of time one may obtain
the Weibull distribution (Mann, Schafer, Singpurivalla, 1974). This
distribution will represent cases where the mortality intensity changes
with time and it has been shown by Pike (1966), Peto and Lee (1973)
and Murthy (1974) that experiments of the type conducted here can
best be analysed by fitting an appropriate Weibull distribution. Upon
this basis a three-parameter Weibull distribution was fitted to the
mortality data from these experiments and used to compare the survival
of the various test groups of salmon. It is also possible, within the
context of the analysis of a Weibull distribution, to accomodate the
case where there is Type I censoring of the data, in which the test
was terminated at a specified time before all fish had died. In this
case the number of deaths and the times-to-death are random variates.
Also the case where there is progressive censoring or withdrawls of
live fish from the test throughout the duration of the test can be
analysed.
The probability density function for the three-parameter Weibull dis-
tribution is given by
* (t - G)13-1 exp [- I (t - G)b]
f (t) = b, c 0; t G 0
0 t G
Where t is the time to death.
The parameter G is sometimes called the threshold parameter since a
death occurs before time G with probability 0. The parameter C usually
27
-------
called the characteristic life, is a scale parameter It is the
100 x [l-exp (-l)]th = 63.2 perceritile of the distribution for any
value of the shape parameter. The parameter b determines the shape
of the distribution. If b=l, the distribution is unimodal and the
ratio of mortality is increasing with time. Further information on
the properties of the Weibull distribution is given in Mann (op« cit.)
and Murthy (op. cit.).
The parameters of the Weibull distribution were estimated using
maximum likelihood procedures as outlined in Mann (op. cit.). Table 4
lists the estimates of these parameters for each test. This distribu-
tion was also used to calculate the probability of survival to a
specified time, t = 60 days, for each test condition (Table 4). The
estimation formula is:
Rt = exp [- (t/c)b].
The expected life (days survival) corresponding to a 25% survival
proportion was estimated by the formula:
P * c [log (I/.25)] 1/b
.25
and the results for each test are shown in Table 4.
28
-------
Table 4. —Estimated values for model parameters., 60-day survival probability and expected life for a
2556 survival proportion.
MODEL PARAMETERS
Type
of Test
Tank No.
Shallow 1
2
3
4
5
6
7
8
9
10
Deep 11
12
13
14
15
16
17
18
19
20
T.D.O.
120
120
115
115
110
109
104
104
100
100
101
104
110
110
116
115
119
119
124
127
Threshold
(G)
0.5
2.4
4.6
6.5
3,9
6.8
0.5
12.8
5.5
8.9
21.7
22.1
12.0
25.0
2.9
2.8
5-9
12.6
2.5
2.5
60-day
survival
Shape Characteristic Life probabilit
(b) {C) (days) (Rgo)
1.335
1.335
1.587
1.088
1.607
1.169
1.107
1.054
1.506
1.151
1.039
1.337
1.517
1.406
1.112
1.376
1.216
1.955
1.450
1.357
23.4
30.1
46.1
47.9
107.9
156.2
368.5
371.0
221.2
491.7
831.6
496.1
384.0
414.4
211.6
315.7
151.3
169.4
40.8
40.0
0.03
0.08
0.22
0.28
0.68
0.72
0.87
0.86
0.87
0.92
0.94
0.95
0.94
0.94
0.78
0.90
0.72
0.88
0.17
0.18
Expected life for
a 25$ survival
,y proportion
(P. 25) " (days)
9,2
11.8
21.0
15.3
49-7
53-8
119.6
113.8
96.7
166.6
250.7
200.7
168.9
170.8
69.0
127.7
54.3
89.6
17.3
16.0
-------
SECTION VI
DISCUSSION
The mortality curves (Fig. 2 & 3) may be affected synergistically
by infection of fish with C. psychrophila after day 64, however,
incidence rate and types of signs of gas-bubble disease on dead
fish showed no apparent difference between the first 64 days and
the last. 63 days — indicating that the effect of _C. psychrophila
was not large. Signs which we attribute to the death of the test
fish (heart and branchial artery embolism) were observed in about
the same percentage of dead individuals before day 64 as after that
time. We, therefore, assume the mortality curves (adjusted for control
mortality) are generally representative of death rates caused by
gas-bubble disease at the dissolved gas concentrations indicated.
The first 64 days have no qualifications, but the last 63 days may
represent a fish stock with less than normal tolerance to excess
dissolved gas pressure.
As shown in figure 10, the rates of mortality and shape of curves
derived from our studies correlate well with experiments done by
Meekin and Turner (1974), in which they exposed juvenile fall chinook
salmon (about 67, 53 and 40 mm long) to 122% N2 + Ar plus 74% 0
(112% T.D.G.). They showed a definite inverse correlation between
size of fish and time to death in supersaturated conditions. Larger
fish (53 mm, 67 mm) succumbed much more rapidly than 40 mm fish,
tested at the same level of saturation (T.D.G.). Also, this same
trend was shown by Shirahata (1966), testing rainbow trout from
hatching to fry stage. Meekin and Turner (1974) also found that
when testing fish for long period of time in supersaturated water,
fish that died were smaller than the survivors which they concluded
may have been caused by retarded growth of the fish that were overcome
by stress. These same general trends appear in our data: i.e.f 1)
juvenile salmon that die from long term stress due to supersaturation
are smaller than their surviving counterparts and; 2) mortality
rates increased commensurate with aging and growth.
30
-------
Certain observations made during these tests have important implica-
tions relative to the experience of natural and hatchery reared
populations of juvenile salmonids. Most areas in the Columbia River
system where salmon and steelhead trout incubate and develop naturally
are located above dams—which cause high concentration of dissolved
gas by heavy spilling of water—and, would be little affected by
long term-low level supersaturation of water,however; spring chinook
salmon and steelhead trout from tributaries of the Willamette River,
a major tributary of the Columbia, make heavy use of areas below
dams where supersaturation is prevalent. Also water for hatcheries
in some instances is either taken from the river below dams where
spilling occurs or from wells that have high concentrations of dis-
solved gas. In these instances effects of exposure to supersaturation
during the early stages of Iife3 such as the period of incubation?
become important.
Significant changes in tolerance to dissolved gas occur at various
life stages. Data from our bioassays and others—Shirahata (1966)
and Meekin and Turner (1974) indicate that (1) yolk sac fry in water
with low levels of supersaturation sustain injuries which become
fatal as the yolk is nearly absorbed; and (2) after fry have "but-
toned up" tolerance to supersaturation becomes quite high but de-
creases gradually thereafter until time of seaward migration. Equil-
ibration of hatchery water sources is thus extremely important at
certain stages of fish development. Also spillway discharges at
certain times will have more effect on survival of downstream juve-
nile migrants than at others; management policies should consider
the changing effects of these discharges.
31
-------
Seaward migration of juvenile salmon and steelhead trout coincides
with the spring freshet in the lower Columbia and Snake rivers.
Freshet conditions are variable from year to year, but heavy spilling
at dams usually occurs every year for some duration. This creates
high gas concentrations of from 1207<> to 140% of saturation. During
years of high flow, these high concentrations persist throughout
long stretches of the river (650 km and more), resulting in long
term exposure for some stocks of fish. Rates of migration of juvenile
fish indicate that at least 28 days is required for travel from
Little Goose Dam to the Columbia River estuary during the highest
flows (Howard Raymond, personal communication*). Even if some fish
are compensating for supersaturation by sounding, a significant
portion of the population is probably exposed to concentrations
of dissolved gas exceeding 1207o saturation.
Data on depth distribution of migrating juvenile fish within the
Snake River near Lower Monumental Dam (Smith,1974) indicate that
58% of the chinook salmon and 36% of the steelhead trout were in
the upper 3.7 m of the water column. Mean depths for these portions
of the migrating stocks were 1.30 m and 1.33 m respectively. This
would compensate for 14.5 - 14.8% effective saturation, which means
that at concentrations of 135%, of saturation—or more—both stocks
of fish would be exposed to gas concentrations equal to—or above--
120%. of saturation for at least 28 days during periods of high flow.
Our studies have shown that, although both the fall chinook salmon
and steelhead trout tend to remain at greater depths with increasing
levels of supersaturation, they are unable to totally compensate
for dissolved gas concentrations above 120%. Therefore, it is im-
perative that corrective measures to reduce supersaturation be im-
plemented as soon as possible to reduce mortality.
^Raymond, Howard L., Fishery Research Biologist, Natl. Oceanic Atmos.
admin., Natl. Mar. Fish. Serv., Northwest Fish. Center, 2725 Montlake
Boulevard E., Seattle, Wa. 98112.
32
-------
<
I-
o
90 r
60
70
60
50
UJ 40
ID
2
r>
o
30
20
10
NMFS /
42 mm /'
II5%TDG
/ 53 mm
112% TDG
f
67mm
112 % TDG
I O JLb** ' ' "- /O ' U^****^
/ I -'i^-- :^~-
/aTcw—(— i " i i L
10 20 30
20 30 40 50
DAYS OF EXPOSURE
42mm
iO%TDG
1 I I I _J > i
60
70
Fig. 10—Mortality versus time curves for fall chinook salmon
at various size (1*0-67 'nun forklength) when exposed
to dissolved gas at 110, 112, and 115% of saturation
T.D.G. in shallow tanks (6.25 m or less). Tests at
T*D.G. were conducted by Meekin and Turner
33
-------
In tests by Dawley and Ebel (in press) the resistance times of steel-
head trout in shallow water tanks at 115% supersaturation was 400%
longer than in our tests with steelhead at the same saturation level.
Fish in the earlier tests were hatchery reared and smaller than
those used in our study; this is probably the reason for their greater
resistance. This difference in resistance, however, is small compared
to the one to two month differences in resistance times we observed
between fall chinook salmon and steelhead trout. This greater differ-
ence correlated well with data from Meekin and Turner (1974) which
indicated that fall chinook salmon were more tolerant to exposure
to supersaturation than were steelhead trout of comparable size
and age. This same order of ranking was noted by Ebel et al. (1971)
and Dawley and Ebel (in press).
Some prominent signs of gas-bubble disease were observed on dead
chinook salmon in association with certain stages in physical develop-
ment or stress experience. Cutaneous blisters in the buccal cavity
and on the body surface and hemorrhages in and around the eye required
more time to develop than other signs, thus did not exist on mortalities
from the higher test levels because of shorter time duration. Therefore,
incidence rate was lower but is entirely dependent on the duration
of stress. Exophthalmia appeared frequently in the 3rd and 4th months,
similar in incidence to blisters on the head. Blisters at the midline
of the vertical surface occurred frequently on dead salmon taken
from the deep and shallow tanks of the highest saturation levels.
This frequency decreased, however, as testing progressed and by
the 4th month there was no evidence of this sign; it appeared to
be related to recent yolk absorbtion. There was a very low incidence
(40%) of blisters between the fin rays at the highest saturation
levels compared to what had been previously observed by other investi-
gators in tests with larger salmonids of other races and species
(near 100%).
-------
Although a portion of the test groups in the deep tanks remained
at sufficient depth to increase their resistance time, the depth
did not provide sufficient compensation to prevent mortality, parti-
cularly at levels above 120%. The mortality rates and signs of gas-
bubble disease of both species also indicated that less hydrostatic
compensation was derived due to depth disposition than expected
when the mean depth of the fish groups are considered. Thus, indivi-
dual fish apparently moved substantially from the observed mean
depth of the test lot. If this did not occur, the effect of hydro-
static compensation would have resulted in a calculated reduction
in effective supersaturation of 12-16% for salmon and 17-20% for
trout. Since the actual mortality rates indicated that only a 107<,
and 10-15% (chinook, steelhead respectively) reduction occurred,
we assume that the fish were moving randomly about within their
observed vertical distribution.
35
-------
SECTION VII
LITERATURE CITED
L. American Public Health Association, American Water Works and Water
Pollution Control Federation. 1971. Standard methods for the exa-
mination of water and wastewater Thirteenth Edition, Am. Public
Health Assoc. New York, N.Y. 874 p.
2. Beiningen, K. 1,, and W. J. Ebel. 1970. Effect of John Day Dam
on dissolved nitrogen concentrations and salmon in the Columbia
River, 1968. Trans. Am. Fish. Soc. 99:664-671.
3. Blahm, T. H., R. V. McConnell, and G. R. Snyder. 1973. Effect
of gas supersaturated Columbia River water on the survival of juve-
nile salmonids. Natl. Oceanic Atmos. Admin., Natl. Marine Fish
Ser., Environmental Field Station, Prescott, OR. Unpubl. manuscr.
61 p. (Processed.)
4. Bouck, G. R. 1972. Effects of gas supersaturation on sample in
the Columbia River. Paper presented at Ecol. Soc. Am. Symp. Aug.
1972 29 p.
5. Bouck, G. R., G. A. Chapman, P. W. Schneider Jr., and D. G. Stevens
1970 Gas-bubble disease in adult Columbia River sockeye salmon
(Onchorchynchus nerka). Pacific Northwest Laboratory, Federal Water
Quality Administration, Corvallis, OR, June, 1970 (Unpub. Man.)
11 p.
6. Coutant, C. C., and R. G. Genoway- 1968. An exploratory study of
interaction of increased temperature and nitrogen supersaturation
on mortality of adult salmonids. BNWL-1529, Battelle-Northwest,
Richland, Washington.
36
-------
7. Dawley, E. M. and W. J. Ebel. In press. Effects of various
concentrations of dissolved atmospheric gas on juvenile chinook
salmon and steelhead trout. Fish. Bull., U.S.
8. Ebel, W. J. 1969. Supersaturation of nitrogen in the Columbia
River and its effect on salmon and steelhead trout. U.S. Fish
Wildl. Serv., Fish Bull. 68:1-11.
9. Ebel, W. J. 1971. Dissolved nitrogen concentrations in the
Columbia and Snake Rivers in 1970 and their effect on chinook
salmon and steelhead trout. U.S. Dep. Commer., Natl. Oceanic
Atmos. Admin., Natl. Mar, Fish. Serv., NCAA Tech. Rep. NMFS,
SSRF-646. 7 p.
10. Ebel, ¥. J. , E. M. Dawley, and B. H. Monk. 1971. Thermal tolerance
of juvenile Pacific salmon and steelhead trout in relation
to supersaturation of nitrogen gas. Fish. Bull., U.S. 69:833-843.
11. Fickeisen, D. H., J. C. Montgomery and M. J. Schneider. Tolerance
of selected fish species to atmospheric gas supersaturation.
Battelle Northwest Lab., Richland, Wa. Unpublished data presented
at A.F.S. meeting, Orlando, Florida, 1973.
12. Harvey, E. N. and A. C. Cooper. 1962. Origin and treatment
of a supersaturated river water. Int. Pac. Salmon Fish. Comm.,
Prog. Rep. No. 9, 19 p.
13. Mann, N. R., R. E. Schafer and N. D. Singuvwalla. 1974. Methods
for statistical analysis of reliability and life data. John
Wiley & Sons, Inc. New York, N. Y. 564 p.
14. McKee, J. E. and H. W. Wolf. 1963. Water quality criteria.
2nd Edition. Calif. State Water Resources Control Board., Publ.
3-A. 548 p. (reprinted 1973).
37
-------
15. Meekin, T. K., and R. L. Allen. 1974. Nitrogen saturation levels
in the mid-Columbia River, 1965-1971. Wash. Dep. Fish., Tech.
Rep. 12:32-77.
16. Meekin, R. K. and B. K. Turner. 1974. Tolerance of salmonid eggs,
juvenile and squawfish to supersaturated nitrogen. Wash. Dep.
Fish., Tech Rep. 12:78-126.
17. Murthy, V. K. 1974. The general point process. Add!son-Wesley
Publishing Company. Reading, Mass. 604 p.
18. Peto, R. and P. Lee. 1973. Weibull distributions for continuous-
carcinogenesis experiments. Biometrics 29:457-470.
19. Pike, M. C. 1966. A method of analysis of a certain class of
experiments in carcinogenesis. Biometrics 22:142-161.
20. Rucker, R. R., and E. M. Tuttle. 1948. Removal of excess nitrogen
in a hatchery water supply. Prog. Fish-Cult. 10:88-90.
21. Shirahata, S. 1966. Experiments on nitrogen gas disease with
rainbow trout fry. Bull. Freshwater Fish. Res. Lab. 15:197-211.
(In Japanese with English Summary).
22. Smith, J. R. 1974. Distribution of seaward-migrating chinook
salmon and steelhead trout in the Snake River above Lower Monu-
mental Dam. Mar. Fish. Rev. 36(8):42-45.
23. Van Slyke, D. D., and J. M. Neill. 1924. The determination of
gas in blood and other solutions by vacuum extraction and mano-
metric measurement. I. J. Biol. Chem. 61:523-574.
38
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-600/3-76-056
3. RECIPIENT'S ACCESSION-NO.
4 TITLE ANDSUBTITLE
SALMONID BIOASSAY OF SUPERSATURATED DISSOLVED AIR
IN WATER
5. REPORT DATE
July 1976 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
N/A
AUTHOR(S)
Earl Dawley, Bruce Monk, Michael Schiewe,
Frank Ossiander, and W. Ebel
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
National Marine Fisheries Service
Northwest Fisheries Center
2725 Montiake Boulevard East
Seattle, Washington 98112
10. PROGRAM ELEMENT NO.
1BA608
11. CONTRACT/GRANT NO.
EPA-1AG-0155(D)
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Duluth, Minnesota 55804
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Tests were conducted in shallow (0.25 m) and deep (2.5 m) tanks of water at IOC with
concentrations of dissolved atmospheric gas ranging from 100% to 127% of air sat-
uration to determine the lethal and sublethal effects on juvenile fall Chinook salmon
(Oncorhynchus tshawytscha) and steel head trout (Salmo gairdneri).
Fall Chinook salmon(average fork length of 42 mm) were much more resistant to super-
saturation than juvenile steelheas trout (average fork length of 180 mm). Salmon
tested in the shallow tanks at 120% of saturation incurred 50% mortality in 22 days
whereas trout tested at the same level incurred 50% mortality in 30 hours. Signs of
bas bubble disease were noted on dead fish and on subsamples of live fish from deep
water tests at 110% saturation and in shallow water tests at 105% or above. Vertical
distribution of both salmon and trout in the deep tanks appeared to compensate for
about 10% and 10-15% respectively of effective saturation. Average depth of the
fish in deep tanks increased with increased gas concentration. Significant differ-
ences in-growth and condition factor of the salmon and trout were not found between
stressed and control fish during the test period.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
animal physiology, salmon, trout, environ-
ments, gas dynamics, pathophysiology,
water pollution, fresh water
gas bubble disease
supersaturation
Pacific salmon
06/c/f
07/b
13/1/k
2. DISTRIBUTION STATEMENT
Unlimited release
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
49
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
39
•k U.S. MVEMUKNT PRINTING OFFICE; U76- 657-695/5457
------- |