Quantifying the effect of ocean environment on the growth rates and age and length at maturity of a

1
Quantifying the effect of ocean environment on
the growth rates and age and length at maturity
of a California Chinook salmon population.
Brian Wells1, Churchill Grimes1, James
Waldvogel2, John Field1, and Christian Reiss3
1. NOAA Fisheries, Santa Cruz, CA 2. Sea Grant,
Crescent City, CA 3. NOAA Fisheries, La Jolla, CA
2
Ocean-type Chinook Stream-type Chinook
Life history
Ocean Stream FW emigration 0 1 Ocean
distribution coastal GOA and gyre Return age
1 - 5 2 - 6
www.fishbase.org
3
Growth related to large-scale environment
OC Ocean-type Chinook
4
WOC Chinook and Summer MEI
WOC Chinook
FL z-scores
Return Year
- (Summer MEI) Lag 1
5
Objectives and Approach
I. Biological
II. Distribution
III. Defining growth model
IV. Size and age at maturation
6
Objectives and Approach
I. Biological
II. Distribution
III. Defining growth model
IV. Size and age at maturation
7
Smith River Estuary
10 km
Photos California Coastal Records Program
8
Biological
9
Defining the growth model
Model building
GRINHERENT (-0.13 age) (-0.011
age2)1.46 N 612
Observed Predicted
Annulus width
Age
10
Defining the growth model
1983 cohort
1989 cohort
Annulus width residual (mm)
1998 cohort
1993 cohort
Year of growth
11
Objectives and Approach
I. Biological
II. Distribution
III. Defining growth model
IV. Size and age at maturation
12
Distribution
13
Distribution
Distribution of CWT captures with origins from
pink region
Col River
C. Blanco
C. Mendocino
M. Bay
14
Objectives and Approach
I. Biological
II. Distribution
III. Defining growth model A. Environmental
variables B. Model building E. Comparison
across regions
IV. Size and age at maturation
15
Defining the growth model
Environmental variables
SSTspr, sum Upwellingspr, sum Scalarspr,
sum NVelocityspr, sum NPseudospr, sum EPseudospr,
sum Curlspr, sum Mixingspr, sum SLHspr, sum
SSTspr, sum Upwellingspr, sum Scalarspr,
sum NVelocityspr, sum NPseudospr, sum EPseudospr,
sum Curlspr, sum Mixingspr, sum SLHspr, sum
SSTspr, sum Upwellingspr, sum Scalarspr,
sum NVelocityspr, sum NPseudospr,
sum EPseudospr, sum Curlspr, sum Mixingspr,
sum SLHspr, sum
Cape Mendocino
M. Bay
reversed index sign
16
Defining the growth model
Comparison across regions (MSEP values)
Based on the fit of the regression to the data, a
jackknifed variance (e.g. MSEP) can be used to
choose the most appropriate model across regions.
GRage
Environment
Further, we can infer that the model fits best
because the fish are affected most strongly by
the pink region - suggesting that they reside in
that region.
17
Defining the growth model
Comparison across regions (MSEP values)
MSEP calculated from a Jackknife iteration
18
Defining the growth model
Environmental variables
Prin 1 -SLHspr,sum, -NPseudospr, Upwellingspr,
-Scalarsum -- spr, sum upwelling
and wind Prin 2 Scalarspr, Mixingspr,
Upwellingsum, Curlsum --
spr turbulence, sum upwelling and curl Prin 3
EPseudospr, -NVelocityspr,sum, -NPseudospr,
-Mixingsum, SSTsum -- Turbulent storm
winds Prin 4 -Espeudosum, NPseudosum, SSTSum
-- sum storms and
SST Prin 5 Mixingspr, NVelocityspr, -Curlsum,
Scalarspr -- spr turbulence from
northerly winds
Prin 1 -SLHspr,sum, Upwellingspr,sum,
NVelocityspr,um
-- spr, sum upwelling
Prin 2 -Scalarsum, -EPseudosum, -NVelocitysum,
Nvelocityspr, Mixingsum -- sum
turbulence Prin 3 -Mixingspr, -Scalarspr
--
spr turbulence Prin 4 NPseudosum,
-Curlsum,spr, UpwellingSum,
-- sum upwelling and
curl Prin 5 -Curlsum, Upwellingspr
-- spr, sum curl and
upwelling
19
Defining the growth model
Model building the first two years of life.
GR1 (-0.04 prin2) (-0.05 prin4)
(-0.02 prin3)
Strong mixing in spring and summer upwelling and
curl retarded growth. Also, NPseudo winds
associated with warm water retarded growth
GR2 (0.03 prin1) (-0.03 prin4) (0.03
prin5) (0.01 prin2) 0.04
Lower SLH and increased spring upwelling improved
growth, but summer scalar winds retarded growth.
Also, NPseudo winds associated with warm water
retarded growth.
20
Defining the growth model
Comparison across region
Maximum recruitment of anchovy and sardine along
the CA Current occurs at wind speeds 6 m/s
(MacKenzie 2000, Faure and Cury 1998, Parrish et
al. 1998, Cury and Roy 1989).
min is 6.22 scalarspr (any direction) min is
5.83 scalarsum
Similar to our trophic argument, Ainley et al
1993 showed the negative effects of turbulence up
the food chain to sea birds.
Recruitment
6
Scalar m/s
21
Defining the growth model
Model building the second two years of life.
GR3 (0.05 prin1) (0.02 prin3) (-0.04
prin5) (0.01 prin2)-0.02
GR4 (0.05 prin1) (0.03 prin3)
Lower SLH and spring and summer upwelling and
northerly winds are good for growth even when
associated with increased mixing.
By the second and third year of growth the prey
window opens up to much larger prey, therefore,
continued recruitment of larval or juvenile prey
is of less importance to the success of Chinook.
Robinson (2004 JFB) Demonstrates that during
turbulent conditions prey species cluster making
them more available to fisheries. It is a small
leap to suggest they are also more available to a
mobile fish predator like Chinook
22
Defining the growth model
1983 cohort
1989 cohort
Annulus width residual (mm)
1998 cohort
1993 cohort
Year of growth
23
Defining the growth model
Observed Predicted
1983 cohort
1989 cohort
Annulus width residual (mm)
1998 cohort
1993 cohort
Year of growth
24
Objectives and Approach
I. Biological
II. Distribution
III. Defining growth model
IV. Size and age at maturation A. Age at
maturity for a cohort B. Length at age of return
25
Size and age at maturation
Age at maturity for a cohort
Observed Predicted
Proportion maturing after 3 OW
Brood year
Prop 3OW (0.36 prin13) (0.15 prin23)
1.79 N 18, r2 0.40 Spring and summer
upwelling during the third year of growth (the
one prior return) increase the likelihood of
returning after the third winter at sea.
26
Size and age at maturation
Length at age of return
Observed Predicted
Length of 3OW returns
Brood year
L3 (9.45 prin23) (7.14 prin13) (-8.13
prin53) (-2.81 prin33) 878.47 N
443 During the third growth year (the one prior
return) spring upwelling and summer mixing and
curl increase the size at return of 3OW fish.
27
Conclusions

• CWT data, Healey 1991, and MSEP comparisons
  indicate that the distribution of Northern
  California Chinook salmon is in source and
  northern region in early life but more southerly
  as they age.
• During first two years of growth, turbulence and
  warm waters were shown to be bad. During next two
  years, upwelling in southern region was good.
• Under good conditions for growth in the year
  prior return, fish mature at 3OW as opposed to
  older.
• And, under good conditions in the third complete
  year of growth, 3OW Chinook return bigger.

Please note the arm waving with regard to
distribution determination.
28
Conclusions

• CWT data, Healey 1991, and MSEP comparisons
  indicate that the distribution of Northern
  California Chinook salmon is in source and
  northern region in early life but more southerly
  as they age.
• During first two years of growth, turbulence and
  warm waters were shown to be bad. During next two
  years, upwelling in southern region was good.
• Under good conditions for growth in the year
  prior return, fish mature at 3OW as opposed to
  older.
• And, under good conditions in the third complete
  year of growth, 3OW Chinook return bigger.

29
Conclusions

• CWT data, Healey 1991, and MSEP comparisons
  indicate that the distribution of Northern
  California Chinook salmon is in source and
  northern region in early life but more southerly
  as they age.
• During first two years of growth, turbulence and
  warm waters were shown to be bad. During next two
  years, upwelling in southern region was good.
• Under good conditions for growth in the year
  prior return, fish mature at 3OW as opposed to
  older.
• And, under good conditions in the third complete
  year of growth, 3OW Chinook return bigger.

30
Acknowledgements
Funding was provided by the Fisheries and the
Environment Program (FATE) of NOAA
Fisheries Suggestions offered by A. MacCall,
N. Mantua, B. MacFarlane, S. Ralston, M.
Schirripa, G. Watters, and E.J. Dick Sample
collection Pacific Salmon Commission CWT
database Backdrops Chrissann Wells
31
end of the show
32
Defining the growth model
Comparison across region 2. Physical
In southern region scalar winds and mixing are
coupled with northerly winds and upwelling.
According to literature upwelling in this region
can be too much (i.e. Scalar gt 6). BUT,
upwelling was not associated with growth.
Therefore, it is the summer turbulence that is
the culprit with no cost or benefit from
upwelling itself.
In source region scalar winds were not related to
any variables.
33
Defining the growth model
Comparison across regions 2. Physical
In support, the composite turbulence variable
(prin1) is positively related to MEI. In ENSO
years turbulence is bad in southern region. The
component captures that there can be too much
turbulence.
r 0.43, N 25
Turbulence component
MEI spring
source region southern region
34
Defining the growth model
Comparison across regions 3 Statistical
Southern region
Source region
35
Domains
X
During warm conditions (SST up) the downwelling
region has increased production/survival while
the upwelling region has decreased
production/survival and vice versa (Hare and
Mantua 1999).
36
Modeling the relationship between growth of
Pacific salmon and the environment.
John Field, Churchill Grimes, Christian Reiss,
Jim Waldvogel, Me (Brian Wells)
1. Synchrony between the growth of Chinook salmon
and the environment
2. The effects of coastal environmental variation
on California Chinook salmon as revealed by scale
increment analysis
37
Part 1. Synchrony between the growth of Chinook
salmon and the environment
Wells, B.K., C.B. Grimes, J.C. Field, C.S. Reiss.
In press. Covariation between the average lengths
of mature coho (Oncorhynchus kisutch) and Chinook
salmon (O. tshawytscha) and the ocean
environment. Fisheries Oceanography.
38
Coho, Chinook distribution and the representation
of index values
ALPI
BC
Col
MEI
WOC
39
Return FL of Chinook
21,000 19,000 14,000
BC - Puget
Columbia
WOC
Ocean Chinook
FL (mm)
Return Year
40
Hypothetical time series
Environment and FL
Return Yr
41
Correlation matrix Chinook and environment
synchrony
Ocean
PDO
MEI
NOI
sp
su
f
sp
su
f
sp
su
f
w
w
w
ALPI
BC-Puget Col WOC
42
Part 2. The effects of coastal environmental
variation on California Chinook salmon as
revealed by scale increment analysis
43
Size and age at maturation
Length at age of return
1.0
Proportion maturing after 3 OW
Length of 3OW returns
0.5
0.0
Brood year