Predation Studies and ManagementIn the Columbia River Basin
Jim PetersenColumbia River Research LaboratoryWestern Fisheries Research CenterU.S. Geological SurveyCook, Washington
CALFED Workshop on PredationJune 22-23, 2005Tiburon, California
Predators
Dams
Major sources of juvenile salmonid mortality
Northern pikeminnow
MajorPredators
Walleye Smallmouth bass
Salmon
Outline• Predation estimation• Management tools
– Predator control– Dam passage options to minimize predation– Dam removal
• American shad issues in the Columbia• Predation beyond the river system
BritishColumbia
Montana
Wyoming
Oregon
Alberta
Washington
Idaho
UtahNevadaCalifornia
PacificOcean
Columbia River Basin
Measure of predation in John Day Reservoir, 1982-1990
• Consumption rate estimation using a meal turnover method – salmonids per predator per day
• Use of a bone key to identify prey species in gut samples
• Population estimates and size structure• Calculation of monthly “loss” = monthly per capita
predation rate x # predators x size …
Salmonids consumed (thousands)by all predators in John Day Reservoir
0
200
400
600
800
1000
April May June July August
Northern pikeminnow consumed 78% of salmonids
Rieman et al. 1990
Dam forebays andtailraces had highpredation losses
Mid-reservoir areastailraces had lowpredation losses
Model predictions
• 10-20% removal especially of larger predators would decrease loss 50%
• Assumed no compensation after removal ( reproductive, growth, consumption rate)
• Density dependence should be evaluated
Rieman and Beamesderfer 1991
Northern pikminnowmanagement program
• Began system-wide in 1990 (Columbia and lower Snake Rivers)
• Currently consists of a bounty fishery and a tribal fishery near dams.
• 2005 bounty pays $4-8 per fish > 9 inches; plus special tagged fish worth $500
• 267,000 caught in 2004; 2.4 million total removed
• Costs BPA approximately $3.1 million per year
BPA website; various publications
Northern pikminnowmanagement program -
evaluation• 9-16% annual exploitation rate• No detected compensation in growth or
fecundity• No change in diet of other predators,
especially smallmouth bass• “predation on juvenile salmon has been cut
by 25%”
BPA website; various publications
Northern pikminnowmanagement program – caveats
and questions• Predator- and prey-density effects have not
been evaluated sufficiently• Predation rates are highly variable (cv = 72-
140%) and large sample sizes will be necessary to detect changes of <100%
Predator density effects
Samples with > 1 salmonid prey
log Predator density
-1
0
1
log Y = 0.36 - 0.42 * log X
log
Pred
atio
n ra
te
0 1 2
McNary Damtailrace
1983-86
Petersen 2002
0
1
2
Columbia River Wind River Clearwater River
Before hatcheryreleaseAfter hatchery release
Pre
datio
n ra
te
Sample distance 1 25 60(km from release)Salmon released 14.4 2.3 1.1(million)
Collis et al. (1995)Shively et al. (1996)
Prey density effects
Prey density effects
McNary Dam tailrace
0 2 4 6 8 100
1
2
3
4
Juvenile salmonid density
Pre
da
tio
n r
ate
(sa
lmo
n /
d) • Type III functional
response• R2 = 46%• 1983-86• f(density, predator
size, prey size, temperature)
Petersen & DeAngelis (1992)
Outline• Predation estimation• Management tools
– Predator control– Dam passage options to minimize predation– Dam removal
• American shad issues in the Columbia• Predation beyond the river system
A Fish Bypass Systems- Turbine Screens- Collection Channel- Transport Pipe- Outfall location
Problems• Physical injury at screen• Predation at outfalls• Survival sometimes low
AcousticTelemetry
Radio Telemetry
The Studies
The Tools
• Route specific survival• Fish passage efficiency• Passage behavior• Tailrace egress
John Day Dam juvenile bypass system
Powerhouse 2
Powerhouse 1
Spillway
New screen bypass outfall
Old Screen bypass outfall
New surface bypass outfall
Predator Locations
(58 individuals)NPMSMB
(58 individuals)
(69 individuals)
April 25 to August 31N2002
Outline• Predation estimation• Management tools
– Predator control– Dam passage options to minimize predation– Dam removal
• American shad issues in the Columbia• Predation beyond the river system
Dams proposed for breachingon the lower Snake River
#
BonnevilleDam
Lower ColumbiaRiver
Lower SnakeRiver
100 km
Washington
Idaho
Oregon
Snake River
Hell’s CanyonDam
Hell’s CanyonDam
Study areaRiver miles147-172
Free-flowinglower SnakeRiver
ColumbiaRiver
Snake River
Mass of salmon consumed
Northern pikeminnow Smallmouth bass
0
1000
2000
3000
4000
5000
6000
Impounded Unimpounded0
1000
2000
Impounded Unimpounded
3000
4000
Juvenile salmon
consumed
(kg)
Change in predation mortality(northern pikeminnow and
smallmouth bass)
1 0.5
59
10
0
10
20
30
40
50
60
70
Impounded Breached Impounded Breached
Predation-related
mortality (%)
April-May June-August
Smallmouth bassJune - July
Crayfish 42% Other 9%
Non-salmonidFish 24%
Salmonids25%
ImpoundedN=852
Salmonids 4%
Free-flowingN=228
Non-salmonidFish 54%
Crayfish 26%
Other 16%
Petersen et al. 2000
Outline• Predation estimation• Management tools
– Predator control– Dam passage options to minimize predation– Dam removal
• American shad issues in the Columbia• Predation beyond the river system
American shad migrating past Bonneville Dam
Photo courtesy of Jim Seelye
American shad passage at Bonneville Dam
0
1
2
3
4
5
6
1936 1946 1956 1966 1976 1986 1996
Mill
ions
Year
Dam construction
#BonnevilleDam
ColumbiaRiver
100 km
Washington
Idaho
Oregon
SnakeRiver
Percent of adult American shad passing dams on thelower Columbia and Snake rivers in 2000
100 5322
5
<1<1
15
100
Juvenile anadromous fishes(Bonneville Dam, 1997)
0
25
50
75
100
April May June July Aug. Sept. Oct.
Americanshad
SalmonidsDailypassageindex
(% of maximum)
John Day Reservoir
River mile 238.5
9/7/94, 1200 h
John Day Reservoir
River mile 238.4
9/7/94, 2200 h
Day
Species compositionby trawl catch
Hydroacoustic surveys
American shad 100%
N = 38
American shad 97%
N = 215
Night
Juvenilefall chinooksalmon
JuvenileAmerican shad
Predators(northern pikeminnow,smallmouth bass,walleye)
Zooplankton,insects, chironomids,etc.
AdultAmerican shad
Juvenile American shad innorthern pikeminnow diet
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 2500 5000Juvenile shad passage index
• Bonneville Dam tailrace
• Aug.-Sept., 1990-91• N = 127-398
predators, 6 dates• Diet: 78% (1990)
5% (1991)
1990
1991
Petersen et al. (1994)
Pred
atio
n ra
t e ( p
rey/
d)
Potential effects of American shad in the diet ofnorthern pikeminnow in the lower Columbia River
Start size = 1044 g, Bonneville temperature, p = 0.173
1000
1100
1200
1300
1400
April May June July Aug. Sept. Oct. Nov.
40% shad, 50% fish, 10% misc
20% shad, 50% fish, 30% misc
50% fish, 50% misc
50% crust, 50% miscMass(g)
Research needs• Hypothesis 1 – Competition.
– Adult shad feed in the lower river and may be competing with outmigrating juvenile salmon.
– Juvenile shad compete with juvenile fall chinook salmon.• Hypothesis 2 – Predation supplementation
– American shad occur in NPM (completed) and SMB diet.• Hypothesis 4 – Nutrient re-distribution.
– Marine derived nutrients – see Limburg p 136• Hypothesis 3 – Disease transmission.
– American shad may be the vector for transmission of Ichthyophonus to marine herring. (?)
–
Caspian terns
Pacific hake
Gulls
Cormorants
Other predators on juvenile salmon
Emerging Tools
Use of Passive Integrated Transponder (PIT) tags anddevelopment of new antenna
systems
American Shad
American shadAlosa sapidissima
Nativerange
American shadAlosa sapidissima
Nativerange
1871
1926
19161876
1876
18911904
1876Kamchatka ?
Gut contents of northern pikeminnowDiet Categories (%)
All Diet CategoriesFish with food in gut
N = 141
Preyfish OnlyN = 42
Preyfish 25%
Crustaceans 16%
Other Invertebrates 2%
Shad 98%
Sculpin 1%Unidentified 1%
Unidentified 57%
1860 1880 1900 1920 1940 1960 1980 2000Im
poun
dmen
t len
gth
(c
umul
ativ
e km
)
0200400600800
100012001400
Impo
undm
ent v
olum
e (x
109 m
3 ; cu
mul
ativ
e)
0102030405060
Flow
(k
cfs)
100200300400500600
Year1860 1880 1900 1920 1940 1960 1980 2000
Salm
onid
s re
leas
ed
(
mill
ions
)
0
50
100
150
200
2501860 1880 1900 1920 1940 1960 1980 2000
Adul
t Am
eric
an s
had
(m
illio
ns)
0
1
2
3
4C
omm
erci
al s
alm
on la
ndin
gs
(mill
ion
kg)
0
5
10
15
20
25
Length
Volume
Fry
Fingerlings andyearlings
A
D
B
Salmon
American shad
C
Juvenile American shad
80 mm
70 mm
60 mm
50 mm
40 mm
Research needs• Hypothesis 1 – Competition.
– Adult shad feed in the lower river and may be competing with outmigrating juvenile salmon.
– Juvenile shad compete with juvenile fall chinook salmon.• Hypothesis 2 – Predation supplementation
– American shad occur in NPM (completed) and SMB diet.
• Hypothesis 4 – Nutrient re-distribution.– Marine derived nutrients – see Limburg p 136
• Hypothesis 3 – Disease transmission. – American shad may be the vector for transmission of
Ichthyophonus to marine herring. (?)–
John Day Reservoir
River mile 238.5
9/7/94, 1200 h
John Day Reservoir
River mile 238.4
9/7/94, 2200 h
Day
Hydroacoustic surveys
Night
0 2 4 6 8 10 12
0
5
10
15
20
25
30
35
400 2 4 6 8 10 12
0
5
10
15
20
25
30
35
40
Dep
th (m
)
Day Night
Fish / 10,000 m3
Juvenilefall chinooksalmon
JuvenileAmerican shad
Predators(northern pikeminnow,smallmouth bass,walleye)
Zooplankton,insects, chironomids,etc.
Juvenile American shad diet(% occurrence)
0 20 40 60 80 100 Cyclopoid Cop.Calanoid Cop.BosminaDaphniaAlonaLeptodoraSidaMoinaChydorusDipteran LarvaePelecypodaCorophiumIlyocryptus
Juvenile American shad diet(% number)
CyclopoidCalanoidBosminaDaphniaLeptodoraOther
Fishing for American shad in the Bonneville Dam tailrace
Photo courtesy of Jim Seelye
Exploitation
050
100150200250300350400450
1930 1940 1950 1960 1970 1980 1990 2000
Cat
ch (x
100
0)
Commercial Sport Columbia Sport Willamette
Bonneville Dam, Columbia River
Research needs
• Role of American shad in salmon recovery efforts– Competition at adult ladders– Competition with juvenile salmon for food or space– Indirect effects, such as supplements to predators
• Effects of shad on zooplankton community• Effects of shad in the lower Columbia River and the
estuary• Nutrient dynamics• Abundance and effects in other river systems of western
North America
Columbia River Basin Predator Removal
Collaborations
11.21.41.61.8
22.22.42.62.8
3
1 2 3 4 5
Number of reservoir areas
Salm
onid
loss
.
• Pooling data influences overall loss estimates
• Higher predation losses in dam forebaysand tailraces
Petersen (1994)
Top Related