Survival at Sea A. Average Marine Mortality B. Life Cycles of specific populations C. Size-Survival D. Variation among years E. Ocean – Upwelling F. Long-term Trends G. Predators
Species Average survival rate
chum 1.4%
pink 2.8%
Chinook 3.1%
coho 10.4
sockeye 13.1%
Average percent marine survival for wild populations
In any population, there is considerable variation in marine survival
0
5
10
15
20
25
1949
1952
1955
1958
1961
1964
1967
1970
1973
1976
1979
Chi
lko
smol
t - a
dult
surv
ival Wild Chilko Lake sockeye salmon
Stage-specific mortality and the dynamics of Fraser River chum salmon
Adults 387,210 Females 200,444 Eggs per female 3,250 Fry 91,990,000 Egg to fry survival 14.17% Adults 812,893 Fry to adult survival 1.19% Return per spawner 2.16
Beacham and Starr. 1982. Fishery Bulletin 80: 813-825
Stage-specific mortality of wild Lewis River chinook
Eggs per female 5,122 Egg to fry survival 65.3% Fry per female 3,345 Fry to smolt survival 10.05% Smolts per female 336 Smolt to adult survival 2.38 Adults per female 8 Escapement 48.8% Returning spawners per female 3.9
McIsaac, D.O. PhD thesis. University of Washington
Stage-specific mortality of Bella Coola pink salmon
Life history stage Mortality Survivors Fry * 60,000,000 1st 40 days 65% 21,000,000 final 410 days 85% 3,150,000 Fishing * 85% 472,500 50% females * 236,250 2,000 eggs/female * 472,500,000 Egg to fry * 87.3% 60,000,000
R.R. Parker. 1968. J.F.R.B.C. 25: 757-794
* Hypothetical values assumed to balance the cycle
Mortality should be is highest on small salmon, soon after they enter the ocean. As they grow, mortality rates are assumed to
decrease (e.g., Karluk Lake sockeye salmon: Ricker 1976).
0200400600800
10001200140016001800
1 5 9 13 17 21 25
Months at sea
Wei
ght (
g)
0
10
20
30
40
50
60
Num
ber o
f sal
monweightnumber
0123456789
0 20 40 60 80 100
MarchApril
Number of days at large
% m
orta
lity
per d
ay
Estimates of average daily mortality of unfed chum salmon fry from Southeast Alaska
(Wertheimer and Thrower 2007 AFS Symposium 57)
8.1%
1.6% 2.3%
4.1%
5.9%
3.9%
Estimated mortality rate of Columbia River and coastal Oregon coho salmon (1982-1985)
Date Days at sea Millions of salmon
% mortality per day
% mortality per period
May 1 0 35.24
June 17 47 8.50 2.92 75.9
Sep 11 133 3.87 0.91 54.5
Sep 1 488 1.11 0.35 71.3
William Pearcy and Joseph Fisher, Oregon State University
What controls the survival of salmon at sea?
• Attributes of the fish – species – size – entry point – wild vs. hatchery reared – migration date
• Attributes of the environment • Predators
0
5
10
15
20
25
0 50 100 150 200 250
Smolt length (mm)
% m
arin
e su
rviv
al
pink
chum
sockeye
chinook
coho
cutthroat
steelhead
Smolt size and variation in marine survival among species
Geographic variation in marine survival of wild coho salmon
Region Site Years SurvivalSE Alaska Auke Cr. 1980 - 2004 20.3
Berners R. 1990 - 2004 17.8Hugh Smith L. 1984 - 2004 12.9Taku R. 1992 - 2004 12.0
N British Columbia Lachmach R. 1988 - 2003 10.0Zolzap Cr. 1993 - 2004 5.9
S British Columbia Black Cr. 1986 - 2004 7.1Salmon R. 1987 - 2004 8.5
Puget Sound Big Beef Cr. 1979 - 2003 17.2Deschutes R. 1980 - 2004 13.1
WA coast Queets R. 1982 - 2003 5.5Bingham Cr. 1983 - 2004 4.4
Shaul et al. 2007 NPAFC Bull. 4: 93-104
Smolt size South (< 55 N)
Middle (56–60 N)
North (> 60 N)
Small (71-76 mm) 5.9 16.7 17.7
Medium (97 mm) 12.6 25.1 32.0
Large (134-139 mm) 17.1 37.0 39.1
Interaction between smolt size and latitude on survival of sockeye salmon: Bigger is
better, and northerly populations do better
(Koenings et al. 1993)
Salmon smolts vary in size among years.
Does size affect their likelihood of survival?
Henderson and Cass. 1991. CJFAS.
Example:
Chilko Lake sockeye salmon:
length distributions among years
Mean smolt fork length was not related to marine survival variation among years
Mean length (mm)
Surv
ival
(%)
72 82 92 102
0
1
0
20
30 Chilko Lake
sockeye salmon
Salmon smolts vary in size within years.
Does size affect their likelihood of survival?
Henderson and Cass. 1991. CJFAS.
Chilko Lake sockeye salmon
Growth History
Scale Radius
Fish
Len
gth Individual circuli
Relationship between fish length and scale radius
0 1 2 3 4 5
100 90 80 70 60 50 40 30 20 10 0
Scale Radius (mm)
Fork
Len
gth
(cm
)
Ishida and Mathews 1989.
Within a year, larger Chilko Lake sockeye salmon smolts were more likely to survive than smaller ones
Surv
ival
(%)
70 80 90 100
0
5
10
1
5
20
Length (mm)
Year 1 Year 2
0
2
4
6
8
10
12
14
0 5 10 15 20 25
Smolt weight (g)
% m
arin
e su
rviv
al
SoosUW
chinook
Site Weight % survival
UW 12.9 2.2
Soos 5.3 0.8
Size was related to survival of hatchery ocean-type Chinook salmon between hatcheries,
but not among years
So, body size does not strongly affect survival among years, but it does seem to play a role in variation in survival within years.
But, for a hatchery to release larger fish, they may hold them longer. What is the interaction between size and release date in controlling survival of salmon going to sea in a given year?
Survival of coho salmon from three size groups, differing from each other by 5 g
0
3
6
9
12
15
20-Apr 10-May 30-May 19-Jun
Date of release
Mar
ine
surv
ival
(%)
smallmediumlarge
Bilton et al. 1984; Morley et al. 1988. Quinsam River Hatchery.
Release date affected survival more than size. Why?
Smolt physiology? Ocean conditions?
Survival of Chinook salmon from the Columbia River declined later in the season, though the patterns varied among years (Scheuerell et al. 2009).
Other evidence that date may play a large role in survival at sea.
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0 0.05 0.1 0.15 0.2 0.25
Average survival of wild coho (5 rivers)
Aver
age
surv
ival
of h
atch
ery
coho
(20
popu
latio
ns)
1:1 ratio
In a given year, wild and hatchery survival rates were correlated, but wild coho salmon had
consistently higher survival rates in Washington (Beetz 2009)
Sometimes there are trends in survival (e.g., stream-type Chinook salmon smolts from the Snake River.
0
1
2
3
4
5
6
7
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
% s
mol
t to
adul
t ret
urn hatchery
wild
(Raymond 1988)
0.00
0.05
0.10
0.15
0.20
0.25
1970 1975 1980 1985 1990 1995 2000 2005
WildHatchery
Prop
ortio
n su
rviv
ing
at s
ea
Washington coho salmon also show trends over time, and the superior performance of wild fish
Beetz 2009
For a long time, many salmon biologists viewed the ocean as:
1) Not limiting salmon production
2) Not changing in any important ways
3) Not knowable anyway
But, in the early 1980s, variation in marine survival made people re-think all this…
Marine survival of Oregon hatchery coho salmon was positively related to an index of upwelling off the coast
0 100 200 300 400 500
0
2
4
6 8
1
0
12
Post-smolt survival (%)
Upwelling S. Mathews (1984)
Comparison of pink salmon caught in the Gulf of Alaska with coho catches in Washington, Oregon
and California (Francis and Sibley, updated)
0
20
40
60
80
100
120
1930 1940 1950 1960 1970 1980 1990
Pink
sal
mon
(mill
ions
)
0
1
2
3
4
5
6
Coh
o sa
lmon
(mill
ions
)
Gulf of Alaska pink salmon
WA, OR, CA coho salmon
Pink salmon catches in central and southeast Alaska seem to follow winter temperatures (November - March air
temperatures in Sitka, southeast Alaska).
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
1930 1940 1950 1960 1970 1980 1990
Win
ter a
ir te
mpe
ratu
re
0
20
40
60
80
100
120
Mill
ions
of s
alm
on
temperaturepink salmon
We now understand that climate and ocean conditions are linked. There are “regimes” that favor salmon in
one part of the range but not others
High survival rates in the south, low in the north
High survival rates in the north, low in the south
Commercial catches of all salmon species in Alaska
0
50,000
100,000
150,000
200,000
250,000
1900 1925 1950 1975 2000
Cat
ches
(100
0s) o
f sal
mon
(Alaska Dept. of Fish and Game data)
Noakes et al. 1997. NPAFC Bull.
All nations, pink, chum, and sockeye catch
Yes, but what do the fish actually die of?
Presumably, the proximate cause of death is normally predation, though the fish may be weakened by disease, lack of stored energy, or poor seawater adaptation, and the vulnerability to predation may be extended if growth is slow.
But, who eats them?
There is no coastwide “enemy of salmon” but rather losses to many different species, depending on the size of the salmon and the region. Overall, little progress has been made on this issue.
Predators: Are harbor seals (for example) a significant predator on salmon?
(Olesiuk 1993)
How many seals are there? 12,990 prior to pupping and 15,810 afterwards.
How much fish do they eat? 1.9 kg or 4.3% body weight per day per seal
What is the composition of their prey?
Mostly Pacific hake from April to November and mostly Pacific herring from December to March in the Strait of Georgia
Harbor seal diet in the Strait of Georgia, B. C.
species tonnes % of total Pacific hake 4214 42.6 Pacific herring 3206 32.4 Pacific salmon 398 4.0 Midshipman 335 3.4 Lingcod 294 3.0 Other 1445 14.6 Total 9892 100%
Estimated mean smolt fork length of Chilko Lake sockeye salmon as outmigrants and surviving adults
Year Outmigrants Adults Probability
1971 81.7 (294) 82.3 (291) < 0.05
1976 88.8 (194) 89.2 (280) > 0.05
1977 95.7 (192) 97.3 (292) < 0.01
In general, fish that survived to adulthood had been larger as smolts than those that died, in a given year (Henderson and Cass 1991).
Life History Stage pink chum sockeye coho chinook steelhead cutthroat
female length (mm) 522 683 553 643 871 721 413
fecundity 1648 2876 3654 2878 5401 4923 1197
egg size (mg) 190 290 130 220 300 150 110
egg to fry survival 0.12 0.13 0.13 0.25 0.38 0.29
fry size (mm) 32 34 28 33 35 28 25
fry to smolt survival 0.26 0.17 0.10 0.14
smolt size (mm) 32 40 80 105 60-120 200 190
smolt to adult survival 0.03 0.01 0.1 0.1 0.03 0.1 0.2
adults per female 5.2 5.1 15.7 12.5 6.4 25.5
egg to smolt survival 0.02 0.02 0.10 0.01
adults per female 10.2 5.4 17.5 9.2
Sockeye salmon marine survival Koenings et al. 1993 CJFAS
70 90 110 130 150 170 Mean Smolt Length (mm)
Smol
t to
Adul
t Sur
viva
l (%
)
50 30 20 10
5 3 2 1