linking marine and freshwater growth in … marine and freshwater growth in western alaska chinook...

Journal of Fish Biology (2009) 75, 1287–1301

doi:10.1111/j.1095-8649.2009.02364.x, available online at www.interscience.wiley.com

Linking marine and freshwater growth in western AlaskaChinook salmon Oncorhynchus tshawytscha

G. T. Ruggerone*†, J. L. Nielsen‡ and B. A. Agler§

*Natural Resources Consultants, 4039 21st Avenue West, Suite 404, Seattle, WA 98199,U.S.A., ‡U.S. Geological Survey, Alaska Science Center, 4210 University Drive, Anchorage,

AK 99508, U.S.A. and §Alaska Department of Fish and Game, Division of CommercialFisheries, Mark, Tag, and Age Lab, 10107 Bentwood Place, Juneau AK 99801, U.S.A.

(Received 4 December 2008, Accepted 28 May 2009)

The hypothesis that growth in Pacific salmon Oncorhynchus spp. is dependent on previous growthwas tested using annual scale growth measurements of wild Chinook salmon Oncorhynchustshawytscha returning to the Yukon and Kuskokwim Rivers, Alaska, from 1964 to 2004. First-year marine growth in individual O. tshawytscha was significantly correlated with growth in freshwater. Furthermore, growth during each of 3 or 4 years at sea was related to growth during theprevious year. The magnitude of the growth response to the previous year’s growth was greaterwhen mean year-class growth during the previous year was relatively low. Length (eye to tail fork,LETF) of adult O. tshawytscha was correlated with cumulative scale growth after the first yearat sea. Adult LETF was also weakly correlated with scale growth that occurred during freshwaterresidence 4 to 5 years earlier, indicating the importance of growth in fresh water. Positive growthresponse to previous growth in O. tshawytscha was probably related to piscivorous diet and foragingbenefits of large body size. Faster growth among O. tshawytscha year classes that initially grewslowly may reflect high mortality in slow growing fish and subsequent compensatory growth insurvivors. Oncorhynchus tshawytscha in this study exhibited complex growth patterns showing apositive relationship with previous growth and a possible compensatory response to environmentalfactors affecting growth of the age class. © 2009 The Authors

Journal compilation © 2009 The Fisheries Society of the British Isles

Key words: Bering Sea; compensatory growth; enhanced growth; salmon recovery; scales.

INTRODUCTION

Growth in Pacific salmon Oncorhynchus spp. has a strong influence on survivalto maturity, life-history characteristics and reproductive success (Beamish et al .,2004; Vøllestad et al ., 2004; Quinn, 2005). Greater growth of Oncorhynchus spp.in fresh water (Henderson & Cass, 1991; Koenings et al ., 1993), early marine life(Ruggerone & Goetz, 2004; Moss et al ., 2005; Farley et al ., 2007) and during thesecond year at sea (Ruggerone et al ., 2003) has been associated with greater survival.Mechanisms linking growth to survival involve higher rates of predation on smallerslow growing individuals and higher lipid content of faster growing individuals thatmay enable Oncorhynchus spp. to survive through the winter when prey are lessavailable (Beamish & Mahnken, 2001). Outside of harvest, however, the cause of

†Author to whom correspondence should be addressed. Tel.: +1 206 285 3480; fax: +1 206 283 8263;email: [email protected]

1287© 2009 The AuthorsJournal compilation © 2009 The Fisheries Society of the British Isles

1288 G . T. RU G G E RO N E E T A L .

death at sea is rarely quantified. Growth also affects key salmonid life-history traits,such as age at maturation, fecundity and egg size (Healey, 1986; Jonsson et al ., 1996;Quinn et al ., 2004). Larger Oncorhynchus spp. have greater competitive ability dur-ing mating, although large body size is disadvantageous in shallow streams (Blairet al ., 1993; Hamon & Foote, 2005). The evolution of anadromy in Oncorhynchusspp. is probably a response to increased growth and associated benefits gained bymigrating to sea where numerous large prey can be found (Gross et al ., 1988; Helle,1989; Quinn & Myers, 2005).

Genetic and environmental factors influence the growth rates of Oncorhynchusspp. In nature, some fishes may encounter periods of low prey availability or lowtemperatures leading to reduced growth. Compensatory growth is a phase of accel-erated growth when favourable conditions are restored after a period of depressedgrowth (Ali et al ., 2003). Although reduced growth may lead to higher risks ofpredation and mortality, a variety of studies indicate individual fishes can undergocompensatory growth such that their growth rate exceeds that of fishes that wereinitially larger (Bilton & Robins, 1973; Damsgard & Dill, 1998; Ali et al ., 2003).

Salmonids can exhibit enhanced growth at different life stages, such as during anontogenetic shift in diet from zooplankton to forage fishes and squid (Kaeriyamaet al ., 2000; Kelley, 2001). At higher latitudes, fish growth at sea is typically fasterthan growth in fresh water (Gross, 1987; Vøllestad et al ., 2004). Faster growingfishes may be more capable of capturing larger prey at an earlier age, leading togreater metabolic efficiency, increased growth and higher survival rates (Mortensenet al ., 2000; Quinn, 2005). Timing of an ontogenetic shift in diet among pinkOncorhynchus gorbuscha (Walbaum) and sockeye salmon Oncorhynchus nerka (Wal-baum) in the North Pacific Ocean may be important for subsequent growth (Aydin,2000, Aydin et al ., 2005). Salminen (1997, 2002) reported that large Atlantic salmonSalmo salar L. smolts experienced enhanced growth relative to small members ofthe cohort because large salmon had greater access to large and energy-rich herringClupea harengus L.

The hypotheses of compensatory, enhanced or random growth responses of fishesto previous growth were tested among Chinook salmon Oncorhynchus tshawytscha(Walbaum). These tests examined whether O. tshawytscha growth at sea was depen-dent on preceding growth in fresh water and the extent to which growth dependencycontinued through each of 3 or 4 years of marine residence. The approach utilizeda long-term collection of scales, which provide an index of growth during each lifestage (Clutter & Whitesel, 1956; Bilton, 1975; Fukuwaka & Kaeriyama, 1997; Fisher& Pearcy, 2005). Growth of individual O. tshawytscha during each year or life stage,based on scale annuli, was compared with growth during the previous year. A pos-itive correlation within each year class of O. tshawytscha would indicate growthenhancement, a negative correlation would suggest growth compensation and a lackof correlation would suggest random growth.

MATERIALS AND METHODS

S C A L E C O L L E C T I O N A N D M E A S U R E M E N T S

The study utilized scales from adult wild O. tshawytscha returning to the Yukon (64◦ 00′N; 164◦ 14′ W) and Kuskokwim Rivers (59◦ 58′ N; 162◦ 26′ W), Alaska, from 1964 to 2004.

© 2009 The AuthorsJournal compilation © 2009 The Fisheries Society of the British Isles, Journal of Fish Biology 2009, 75, 1287–1301

G ROW T H O F O N C O R H Y N C H U S T S H AW Y T S C H A 1289

These catchments, the largest in Alaska, have supported annual O. tshawytscha harvests (com-mercial and subsistence) averaging 147 000 and 99 000 fish, respectively, since 1961 (Bue &Hayes, 2006; Whitmore et al ., 2008). Scales were obtained from the Alaska Department ofFish and Game (ADFG) archive in Anchorage, Alaska (http://www.adfg.state.ak.us/). For theYukon River, scales were selected for measurement only when they were from O. tshawytschacaptured with set gillnets (21·6 cm stretched mesh) located in the lower river (river km20–30). Fewer scales were available from fish harvested in the Kuskokwim River; therefore,these scales were selected from multiple gear types. In most years, Kuskokwim O. tshawytschascales were selected from fish captured by gillnets (15·2 or 21·6 cm mesh) in the mainstemriver near Bethel, but some scales were sampled at tributary weirs. The scales were typicallyobtained from one gear type within a given year. In both catchments, scales were primarilycollected from June to July in an attempt to consistently select fish from the same stocks.

Approximately 50 scales from each of the two dominant O. tshawytscha age groups fromboth the Yukon and Kuskokwim Rivers were measured, i.e. c. 200 scales per year. These fishspent one winter in fresh water (i.e. river-type fish) and three (age 1·3) or four winters (age1·4) in the ocean. Scales were selected for measurement only when: (1) the age determina-tion agreed with the one previously made by ADFG, (2) the scale shape indicated the scalewas removed from the preferred area (Koo, 1962) and (3) the circuli and annuli were clearlydefined and not affected by scale regeneration or significant resorption along the measurementaxis.

Scale measurements followed procedures described by Hagen et al . (2001). After select-ing a scale for measurement, the scale was scanned from a microfiche reader and stored asa high resolution digital file. The high resolution image (3352 × 4425 pixels) allowed theentire scale to be viewed and provided enough pixels between narrow circuli to ensure accu-rate measurements of circuli spacing. The digital image was loaded in Optimas 6·5 imageprocessing software (www.mediacy.com) to collect measurement data using a customizedprogramme. The scale image was displayed on an LCD monitor, and the scale measurementaxis was defined as the longest axis extending from the scale focus. Distance (mm) betweencirculi was measured within each growth zone, i.e. growth through the first winter in freshwater from the scale focus to the outer circulus of the first freshwater annulus (FW1), springplus growth zone (FWPL) extending from the last FW1 circulus to the first marine circulus(smolt transition period), each annual ocean growth zone (SW1, SW2, SW3 and SW4), andfrom the last ocean annulus to the edge of the scale (SWPL). Data associated with each scale,such as date of collection, location, sex, fish length (mid-eye to tail fork, LETF) and capturemethod, were included in the dataset.

DATA A NA LY S I S

Linear regressions were performed to determine whether scale growth during each lifestage (FWPL, SW1, SW2, SW3, SW4 and SWPL) could be explained by scale growth duringthe previous stage. Regressions were performed annually for scale growth measurements ofindividual fish from each stock (Yukon and Kuskokwim) and each age group (ages 1·3 and1·4). Sample groups having <20 scale measurements were excluded from regression analysis.Exploratory analyses indicated that the best explanatory variable was the distance across thefive widest (adjacent) circuli during the previous stage (four circuli in FW1). Total annual scalegrowth could have also been used as the primary explanatory variable. It yielded comparableresults, but peak circuli growth consistently provided the highest correlation. When comparinggrowth during the smolt transition period (FWPL) with growth in fresh water, the entire FW1scale growth was used because juvenile size has been shown to affect timing of Oncorhynchusspp. smolt migration and therefore growth during this brief transition period (Burgner, 1962).

It was hypothesized that growth of individual O. tshawytscha during the past four decadeswas dependent on the previous year’s growth. A t-test was used to examine whether regressionslopes (mm per mm of scale growth) calculated for each year class, age group, stock and lifestage were statistically different from zero (α = 0·05).

Growth of O. tshawytscha during each life stage was expected to vary from year to year.Therefore, additional regressions were calculated to determine whether the regression slopes(scale growth v. previous year’s growth) were related to mean scale size of the year class



during the previous life stage. In other words, was the magnitude of the growth responsedependent on prior mean scale growth of the fish year class? Scale growth values were basedon the mean of ages 1·3 and 1·4 fish. Mean scale size was calculated from the mean ofnormalized values (s.d. > or < the long-term mean) for males and females, which mayexperience differential growth. This analysis only involved Yukon River fish because scalesfrom Yukon River fish were collected consistently across all years (same location, gear andmesh size), thereby minimizing possible bias in annual mean values associated with differentsampling methods. The Durbin–Watson statistic (Kutner et al ., 2005) was used to test forautocorrelation among regression residuals. If the presence of autocorrelation was significantor inconclusive, then the slope of the previous year class was included in a multiple regressionas a means to remove autocorrelation. Partial residual analysis (Larsen & McCleary, 1972)was used to show the relationship between the regression slope of a year class and the growthduring the previous year after removing the autocorrelation effect.

The adult LETF of individual O. tshawytscha (all stocks and ages combined) was comparedwith their scale measurements using regression analysis to examine whether scale growthwas related to LETF. Exploratory analyses indicated a variety of scale measurements werecorrelated with adult LETF, but the analysis presented here was based on cumulative scalegrowth from the second year at sea (SW2) through to the third circulus of the homewardmigration period (SWPL). The outer SWPL zone was excluded due to scale resorption beyondthe third circuli in some scales.

Correlation between adult LETF and scale growth in fresh water 4 to 5 years earlier wasexpected to be weak. Correlation coefficients were calculated for each stock, age group andyear of adult return. Two methods were used to test for positive correlation across all years.First, a χ2 test using Yates correction for continuity was used to test the null hypothesis thatthe frequency of positive correlations across the past four decades was random (i.e. 0·5), asexpected if growth in fresh water had no effect on adult LETF. Second, a t-test was used toevaluate whether the mean annual correlation coefficient during the past four decades was >0.

RESULTS

G ROW T H R E L AT I O N S H I P S W I T H I N I N D I V I D UA LO N C O R H Y N C H U S T S H AW Y T S C H A

A total of 690 regressions were performed to evaluate the relationship betweenannual scale growth of individual O. tshawytscha and previous year’s growth foreach year of adult return from 1964 to 2004. Age 1·4 scales from O. tshawytschareturning to the Yukon River in 1991 provided an example of these regressions duringeach life stage (Fig. 1). Annual growth during each life stage at sea was correlatedwith the previous year’s growth (P < 0·001), with one exception. Annual growthduring the smolt transition from fresh to marine waters (brief transition and minimalgrowth) was not correlated with freshwater growth (P > 0·05).

On average, 48% of the 131 regressions of early marine growth (SW1) on freshwa-ter growth (FW1) were positively correlated and statistically significant (P < 0·05),indicating that growth during the first year at sea was related to growth during fresh-water residence (Table I). This pattern held for each stock and age group. Negativecorrelations represented only 15% of the total, and none were statistically significant.The variability in early marine growth explained by peak growth in fresh water was13%, on average, but it reached as much as 55% for age 1·3 fish returning to theKuskokwim River in 1975 (Table I).

The relationship of scale growth to the previous year’s growth increased duringlater life stages in the ocean. On average, 71% of the 131 regressions of SW2 on SW1growth had a positive slope and were statistically significant (P < 0·05), and 98% of



Tab

leI.

Perc

enta

geof

annu

allif

e-st

age

grow

thre

gres

sion

sfo

rea

chst

ock

and

age

grou

pof

Onc

orhy

nchu

sts

haw

ytsc

hadu

ring

the

past

thre

eto

four

deca

des

that

wer

est

atis

tical

lysi

gnifi

cant

(P<

0·05)

.Dep

ende

ntva

riab

les

wer

eto

tal

scal

egr

owth

duri

ngea

chlif

est

age,

and

inde

pend

ent

vari

able

sw

ere

eith

erpe

akor

tota

l(F

WPL

onFW

1on

ly)

scal

egr

owth

duri

ngth

epr

evio

uslif

est

age.

The

mea

npe

rcen

tage

ofan

nual

regr

essi

ons,

incl

udin

gno

n-si

gnifi

cant

regr

essi

ons,

havi

nga

posi

tive

slop

ean

dth

em

ean

and

max

imum

annu

alco

rrel

atio

nco

effic

ient

(r2)

for

each

life

stag

ear

esh

own

Reg

ress

ion

Num

ber

ofA

vera

gesc

ales

FWPL

SW1

SW2

SW3

SW4

SWPL

onSt

ock

&A

geye

ars

per

year

onFW

1on

FW1

onSW

1on

SW2

onSW

3SW

3or

SW4

Yuk

on(1

966

–20

04)

Age

1·334

42·7

0%47

%79

%62

%N

A95

%A

ge1·4

3950

·013

%67

%79

%92

%72

%89

%K

usko

kwim

(196

4–

2004

)A

ge1·3

2937

·03%

38%

52%

41%

NA

76%

Age

1·429

40·7

21%

41%

72%

59%

48%

100%

Mea

n±

s.d.

%si

gnifi

cant

regr

essi

ons

9±

9%48

±13

%71

±13

%64

±21

%60

±17

%90

±10

%M

ean

±1

s.d.

%po

sitiv

esl

opes

41±

14%

85±

12%

98±

3%96

±2%

97±

4%99

±2%

Mea

n±

s.d.

coef

ficie

ntof

dete

rmin

atio

n(r

2)

0·05

±0·0

10·1

3±

0·03

0·22

±0·0

40·1

9±

0·07

0·15

±0·0

30·4

0±

0·03

Max

imum

coef

ficie

ntof

dete

rmin

atio

n(r

2)

0·39

0·55

0·81

0·59

0·48

0·73

FWPL

,sc

ale

grow

thdu

ring

the

smol

ttr

ansi

tion

peri

odin

spri

ngfr

omfr

esh

wat

erto

the

ocea

n;FW

1,sc

ale

grow

thdu

ring

the

first

year

infr

esh

wat

er;

SW1,

scal

egr

owth

duri

ngth

efir

stye

arat

sea;

SW2,

scal

egr

owth

duri

ngth

ese

cond

year

atse

a;SW

3,sc

ale

grow

thdu

ring

the

thir

dye

arat

sea;

SW4,

scal

egr

owth

duri

ngth

efo

urth

year

atse

a;SW

PL,

scal

egr

owth

duri

ngth

eho

mew

ard

mig

rati

on.



0·0

0·1

0·2

0·3

0·12 0·16 0·20 0·24 0·28 0·32 0·36

FWPL

(m

m)

FW1 (mm)

r2 = 0·01

0·6

0·8

1·0

1·2

1·4

1·6

0·04 0·06 0·08 0·10 0·12 0·14

SW1

(mm

)

FW1 (mm)

r2 = 0·36

0·5

0·7

0·9

1·1

1·3

0·20 0·24 0·28 0·32 0.36 0.40

SW2

(mm

)

SW1 (mm)

r2 = 0·45

0·4

0·6

0·8

1·0

1·2

1·4

0·15 0·20 0·25 0·30 0·35

SW3

(mm

)

SW2 (mm)

r2 = 0·38

0·4

0·6

0·8

1·0

1·2

0·15 0·20 0·25 0·30 0·35 0·40

SW4

(mm

)

SW3 (mm)

r2 = 0·27

0·06

0·08

0·10

0·12

0·14

0·16

0·18

0·20

0·15 0·20 0·25 0·30 0·35 0·40

SWPL

(m

m)

SW4 (mm)

r2 = 0·69

Fig. 1. Regressions of scale growth during each life stage with growth during the previous year (FWPL,scale growth during transition from fresh water to the ocean; FW1, first year in fresh water; SW1,first year at sea; SW2, second year at sea; SW3, third year at sea; SW4, fourth year at sea; SWPL,homeward migration). Regressions are based on individual age 1·4 Oncorhynchus tshawytscha returningto the Yukon River, Alaska, in 1991. This cohort exhibited relatively high growth correlations. Theindependent variables were peak scale growth during each life stage (except FWPL regressed withannual scale growth during the first year in fresh water, FW1). Regressions were statistically significant(P < 0·001) except for FWPL (P > 0·05).

all correlations were positive. Likewise, 64, 60 and 90% of the correlations involvingthe dependent variables SW3, SW4 and SWPL had a positive slope and were statis-tically significant with previous year’s growth (Table I). The percentage of positivecorrelations ranged from 84 to 99% of the total. On average, 15 to 40% of the vari-ability in marine scale growth was explained by the previous year’s growth, depend-ing on life stage. For some cohorts, previous growth explained up to 48 to 81% of thevariability in marine growth, depending on life stage. The strength of the relationshipstended to be greater for age 1·4 fish rather than age 1·3 fish during each life stage.



In contrast, O. tshawytscha scale growth during FWPL was typically not related togrowth during the previous year in fresh water. Only 9% of the 131 regressions werestatistically significant, including both positive and negative relationships (Table I).The few significant correlations could be due to chance.

G ROW T H R E L AT I O N S H I P S AC RO S S Y E A R S

When comparing across all years, the mean slope of the annual regressions ofgrowth on the previous year’s growth was statistically greater than zero for eachstock, age group and life stage of O. tshawytscha (P < 0·001), except during FWPL(Fig. 2), indicating that growth was related to previous year’s growth throughoutmost life stages during the study period. Mean slope of the regressions of FWPLon freshwater growth during the previous year was significantly <0 for three ofthe four tests (Fig. 2), indicating a unique relationship involving FWPL and FW1growth. Annual regression slopes of most life stages did not vary with time or withocean climate shifts observed in 1977, 1989 and 1997 (Kruse, 1998; Hare & Mantua,2000), except the magnitude of slopes for FWPL (r = 0·66) and SW1 (r = 0·42)were positively correlated with time (P < 0·05).

G ROW T H R E L AT I O N S H I P S R E L AT I V E TO M E A N S C A L EG ROW T H O F T H E Y E A R C L A S S

The relationship of O. tshawytscha scale growth with previous year’s growthwas stronger when previous year’s growth of the population was relatively low,especially during early marine life. Slope of the annual regressions of scale growthon the previous year’s growth was negatively correlated with mean scale growthof the year class during the previous year for all life stages except those duringlate marine life, i.e. SW4 and SWPL (Fig. 3). Autocorrelation was non-significant(P > 0·05) when comparing the SW1 and SW2 slopes with mean growth during theprevious year. The potential effect of autocorrelation, based on inconclusive statisticaltests, on the FWPL and SW3 relationships was removed by inclusion of the slopefrom the previous year class (P < 0·05). In these multiple regression models, thenegative relationship between mean SW2 growth and SW3 slope was significant(partial P < 0·05), whereas the negative relationship between mean growth in freshwater (FW1) and FWPL slope was not significant (P > 0·05).

A D U LT LETF A N D S C A L E G ROW T H

Regressions of individual adult O. tshawytscha LETF and scale measurements(SW2 to SWPL) were statistically significant for 97% of the annual regressions since1964 (P < 0·05). On average, 32% of the variability in adult LETF was explainedby cumulative scale growth during the second year at sea to the homeward migrationperiod (SWPL first three circuli). In some years, adult LETF was highly correlatedwith scale growth (Fig. 4).

Adult LETF was significantly, albeit weakly, related to growth during the freshwa-ter period. Adult LETF of individual age 1·3 and 1·4 Yukon and Kuskokwim fish waspositively correlated with total growth in FW1 for 63 and 64% of the annual tests,respectively. These percentages include statistically significant and non-significant



K 1·3

K 1·4

Y 1·3

Y 1·4

K 1·3

K 1·4

Y 1·3

Y 1·4

K 1·3

K 1·4

Y 1·3

Y 1·4

K 1·3

K 1·4

Y 1·3

Y 1·4

K 1·3

K 1·4

Y 1·3

Y 1·4

K 1·3

K 1·4

Y 1·3

Y 1·4

−2·5 −1·5 −0·5 0·5 1·5 2·5 3·5 4·5

Scale growth slope (mm mm–1)

FWPL on FW1

SW1 on FW1

SW2 on SW1

SW3 on SW2

SW4 on SW3

SWPL onSW3 or SW4

P > 0·05

P < 0·05

P < 0·05

P < 0·01

Age

Old

erY

oung

er

Fig. 2. Mean rate of scale growth change in relation to previous year’s growth during each life stage and ageclass (1·3 and 1·4) of Yukon River (Y) and Kuskokwim River (K) Oncorhynchus tshawytscha. Rate ofscale growth change is mean ± 95% CI slope calculated from annual regressions of scale growth onprevious year’s growth (see Fig. 1). Slope values were statistically significant (P < 0·001) except asnoted.



−3·0−2·5−2·0−1·5−1·0−0·5

0·00·51·01·5

−1·5 −1·0 −0·5 0·0 0·5 1·0 1·5 2·0

FWPL

slo

pe (

mm

mm

–1 )

FW1 (Z )

2·0

r2 = 0·13

−1·5

−1·0

−0·5

0·0

0·5

1·0

1·5

2·0

−1·5 −1·0 −0·5 0·0 0·5 1·0 1·5 2·0

SW1

slop

e (m

m m

m–1 )

FW1 (Z )

r2 = 0·22

−1·5

−1·0

−0·5

0·0

0·5

1·0

1·5

−1·5 −1·0 −0·5 0·0 0·5 1·0 1·5 2·0

SW2

slop

e (m

m m

m–

1 )

SW1 (Z )

r2 = 0·11

−1·5−1·0−0·5

0·00·51·01·52·02·53·0

−2·5−2·0−1·5−1·0−0·5 0·0 0·5 1·0 1·5 2·0

SW3

slop

e (m

m m

m–1 )

SW2 (Z )

r2 = 0·16

Fig. 3. Regressions of normalized annual scale growth slopes of Yukon River Oncorhynchus tshawytscha onnormalized mean scale growth (Z) of the year class during the previous year (see Fig. 1). Scale slopeswere calculated from regressions of scale growth of individual O. tshawytscha on the previous year’sgrowth. The FWPL and SW3 plots were based on partial residuals of multiple regressions that includedthe slope of the previous year class to remove autocorrelation. In these cases, r2 is based on the partialcorrelation coefficient. All correlations were statistically significant (P < 0·05) except the FWPL slope(partial P > 0·05). Regressions involving SW4 and SWPL life stages were non-significant and are notshown (P > 0·05).

correlations. χ2 analysis indicated that the frequency of positive v. negative correla-tions was not random among Yukon River (χ2, n = 73, P < 0·05), Kuskokwim River(χ2, n = 58, P < 0·05) and both stocks combined (χ2, n = 131, P < 0·05). Like-wise, annual correlation coefficients (r) were statistically >0 for both Yukon (t-test,mean r = 0·07, d.f. = 76, P = 0·001) and Kuskokwim (mean r = 0·07, d.f. = 59,P < 0·01) O. tshawytscha.

DISCUSSION

Analysis of Yukon and Kuskokwim O. tshawytscha scale growth within each yearduring the past four decades indicated that annual growth of individuals was relatedto growth during the previous year. This relationship was maintained between freshwater and early marine life stages and among each life stage in the ocean. The degreeof growth relationship, as indicated by the magnitude of the regression slope, wasgreater when mean growth of the year class was relatively low during the previousyear. Adult LETF of O. tshawytscha was most closely correlated with cumulativescale growth after the first year at sea. Adult LETF, however, was also weakly relatedto growth that occurred during freshwater residence 4 to 5 years earlier.



500

600

700

800

900

1000

1100

1·5 2·0 2·5 3·0 3·5 4·0

500

600

700

800

900

1000

1100

1·5 2·0 2·5 3·0 3·5 4·0

Scale radius (mm)

(a)

(b)

LE

TF

(mm

)

Fig. 4. Examples of the relationship between adult length (eye to tail fork, LETF) of individual Yukon RiverOncorhynchus tshawytscha and scale radius during (a) 1980 and (b) 1986. Scale radius is the sum of thesecond, third and fourth years of marine growth (SW2, SW3 and SW4) and the first three circuli duringthe home migration period (SWPL). Age 1·3 and 1·4 fish were included in the regressions. The curveswere fitted by (a) y = 142·2x + 454(r2 = 0·73) and (b) y = 145·6x + 396(r2 = 0·73).

The relationship of individual O. tshawytscha growth to the previous year’s growthmay relate to their piscivorous feeding habits and their tendency to select rela-tively large prey including squid while foraging in the ocean (Major et al ., 1978;Healey, 1991; Farley et al ., 2009). Greater growth may beget greater growth becauselarger Oncorhynchus spp. can consume larger prey (Ruggerone, 1992; Juanes, 1994),thereby enhancing overall availability and energy content of prey for largerOncorhynchus spp. (Salminen, 1997; Aydin et al ., 2005). Somatic growth may beespecially important for O. tshawytscha because they mature at a large size relative toother species of Oncorhynchus spp. Rapid growth and large body size may enhancesurvival (Healey, 1982) and lead to greater fecundity at a given age (Quinn, 2005).



Investigations to determine whether Oncorhynchus spp. smolt size has an effecton post-smolt growth have led to inconsistent findings. Studies on S. salar indicatedgrowth was compensatory or negatively correlated with smolt size (Skilbrei, 1989;Nicieza & Brana, 1993; Einum et al ., 2002), independent of smolt size (Friedlandet al ., 2006) or positively correlated with smolt size (Salminen, 1997). Friedlandet al . (2006) also reported that growth at sea of S. salar during summer was pos-itively correlated with their growth during spring. Post-smolt growth of steelheadOncorhynchus mykiss (Walbaum) was positively correlated with smolt size, whereaspost-smolt growth of coho salmon Oncorhynchus kisutch (Walbaum) was either neg-atively correlated or not correlated with smolt size (Johnsson et al ., 1997; Snoveret al ., 2005). In Bristol Bay, Alaska, scale growth of individual O. nerka dur-ing the second year at sea was not correlated with growth during the first year(G. T. Ruggerone, unpubl. data). The relationship between growth and size maydepend, in part, on whether the fish is consuming relatively large mobile prey versussmaller less mobile prey.

The degree of growth enhancement (magnitude of positive regression slope) amongO. tshawytscha varied from year to year. It was greatest when the mean growth ofthe year class was relatively low indicating the tendency for fish growth to increaseeven more when previous growth was relatively low. This form of compensatorygrowth, or ‘catch-up’ growth, might be related to both physiological and ecologicalfactors. The O. tshawytscha scales used in this study were from fish that survived 3or 4 years at sea. Conceivably, mortality may have been relatively high for slowergrowing individuals, leading to the observed rapid growth increase when body sizeof the population was relatively small. This pattern was also observed among BristolBay O. nerka in which mean growth of the population during the second year at seawas inversely correlated with early marine growth (Ruggerone et al ., 2005).

Oncorhynchus tshawytscha scale growth during the FWPL tended to be nega-tively correlated with growth in fresh water during the previous year, although therelationships were weak. This relationship probably reflects a behavioural response ofO. tshawytscha rather than compensatory growth. Smaller, slow growingOncorhynchus spp. tend to delay the date of smoltification, leading to greater timeand growth during the spring transition period from fresh to marine waters (Burgner,1962). This transition growth was reflected in the FWPL measurements.

Adult LETF of O. tshawytscha was correlated with scale growth, but variabilityexplained by scale measurements was only 32%, on average. The low LETF andscale correlation at the adult stage compared with the juvenile and immature stages(Fukuwaka & Kaeriyama, 1997; Fisher & Pearcy, 2005) probably reflects a change inthe length and scale relationship caused by rapid growth of maturing Oncorhynchusspp., as they migrate toward their natal river (Brett, 1995) followed by reduced feed-ing and resorption of the scale with the onset of maturation (Bilton, 1985; Price &Schreck, 2003). Still, the adult LETF and scale relationship provides evidence thatthe observed scale growth relationships between adjacent life stages were related tobody growth rather than an artifact of scale growth.

Evidence suggests that climate shifts and associated oceanic conditions can influ-ence Oncorhynchus spp. abundance through their effects on growth during earlymarine life (Peyronnet et al ., 2007; Ruggerone et al ., 2007). Abundance of west-ern Alaska O. tshawytscha, including those in the Yukon and Kuskokwim Rivers,increased after the 1977 regime shift, then declined after the 1997 to 1998 El Nino



(Kruse, 1998). The present study did not identify distinct and immediate shifts inO. tshawytscha growth in relation to climate shifts, possibly because O. tshawytschafeed in the ocean at a relatively high trophic level that involves complex speciesinteractions (Hunt et al ., 2002) that may delay a response to climate shifts. Further-more, the present study indicated that growth of O. tshawytscha was influenced, inpart, by growth during earlier life stages, suggesting that effects of climate shifts ongrowth may be mediated by earlier growth.

The dependency of early marine growth and adult LETF of O. tshawytscha ongrowth in fresh water has important implications for maintaining habitat quality infresh water. Oncorhynchus spp. habitat from California to the Pacific Northwest isoften degraded by human activities leading to depleted stocks and protection underthe United States Endangered Species Act (Yoshiyama et al ., 1998; Good et al .,2005). Atmospheric warming is predicted to lead to amplification of precipitationextremes affecting freshwater systems (Allen & Ingram, 2002). Predicted increases inextreme precipitation events correlate with changes in sea surface temperature (Allan& Soden, 2008) and by implication change in the ocean prey resource for salmonids.Although O. tshawytscha may compensate for small size in fresh water, the presentstudy suggests that rapid growth during early life stages, including the freshwaterperiod, is important to subsequent growth and presumably survival throughout life.Thus, growth in freshwater habitats may be especially important during periods oflow ocean productivity because large size at ocean entry may enable fish to consumea wider spectrum of prey.

The manuscript was partially prepared under award NA06FP0387 from the NationalOceanic and Atmospheric Administration, U.S. Department of Commerce, administered bythe Alaska Department of Fish and Game. This investigation was funded by the Arctic-Yukon-Kuskokwim Sustainable Salmon Initiative and the U.S. Geological Survey’s Global ChangeInitiative Program. We thank ADFG biologists, especially D. Molyneaux, who provided accessto numerous reports. We also appreciate efforts to gather, measure and catalogue scales byD. Folletti, M. Lovejoy, A. Norman, D. Oxman, W. Rosky and W. Whelan. Constructivecomments on the manuscript were provided by S. Goodman, P. Hagen, E. Ivaska and twoanonymous reviewers. Use of any trade names or products is for descriptive purposes onlyand does not imply endorsement of the U.S. Government. The statements, findings, conclu-sions and recommendations are those of the authors and do not necessarily reflect the viewsof the National Oceanic and Atmospheric Administration, the U.S. Department of Commerce,the U.S. Department of Interior or the Alaska Department of Fish and Game.

References

Ali, M., Nicieza, A. & Wootton, R. J. (2003). Compensatory growth in fishes: a response togrowth depression. Fish and Fisheries 4, 147–190.

Allan, R. P. & Soden, B. J. (2008). Atmospheric warming and the amplification of precipi-tation extremes. Science 321, 1481–1484.

Allen, M. R. & Ingram, W. J. (2002). Constraints on future changes in climate and the hydro-logic cycle. Nature 419, 224–232.

Aydin, K. Y. (2000). Trophic feedback and carrying capacity of Pacific salmon (Oncorhynchusspp.) on the high seas of the Gulf of Alaska. PhD Thesis, University of Washington,Seattle, WA, USA.

Aydin, K. Y., McFarlane, G. A., King, J. R., Megrey, B. A. & Myers, K. W. (2005). Link-ing oceanic food webs to coastal production and growth rates of Pacific salmon(Oncorhynchus spp.), using models on three scales. Deep Sea Research II 52, 757–780.



Beamish, R. J. & Mahnken, C. (2001). A critical size and period hypothesis to explain nat-ural regulation of salmon abundance and the linkage to climate change. Progress inOceanography 49, 423–437.

Beamish, R. J., Mahnken, C. & Neville, C. M. (2004). Evidence that reduced early marinegrowth is associated with lower marine survival of coho salmon. Transactions of theAmerican Fisheries Society 133, 26–33.

Bilton, H. T. (1975). Factors influencing the formation of scale characters. Bulletin Interna-tional North Pacific Fisheries Commission 32, 102–108.

Bilton, H. T. (1985). The body-scale relationship for adult sockeye (Oncorhynchus nerka),pink (O. gorbuscha), chum salmon (O. keta), and Chinook salmon (O. tshawytscha).Canadian Technical Report of the Fisheries and Aquatic Sciences 1353.

Bilton, H. T. & Robins, G. L. (1973). The effects of starvation and subsequent feeding onsurvival and growth of Fulton Channel sockeye salmon fry (Oncorhynchus nerka).Journal of the Fisheries Research Board of Canada 30, 1–5.

Blair, G. R., Rogers, D. E. & Quinn, T. P. (1993). Variation on life history characteristicsand morphology of sockeye salmon in the Kvichak River system, Bristol Bay, Alaska.Transactions of the American Fisheries Society 122, 550–559.

Brett, J. R. (1995). Energetics. In Physiological Ecology of Pacific Salmon (Groot, C.,Margolis, L., & Clarke, W. C., eds), pp. 1–68. Vancouver, BC: University of BritishColumbia Press.

Bue, F. J. & Hayes, S. J. (2006). 2006 Yukon area subsistence, personal use, and commercialsalmon fisheries outlook and management strategies. Fishery Management Report No.06-32. Anchorage, AK: Alaska Department of Fish and Game, Divisions of Sport Fishand Commercial Fisheries.

Burgner, R. L. (1962). Studies of red salmon smolts from the Wood River Lakes, Alaska. InStudies of Alaska Red Salmon (Koo, T. S. Y., ed.), pp. 247–314. Seattle, WA: Univer-sity of Washington Press.

Clutter, R. I., & Whitesel, L. E. (1956). Collection and interpretation of sockeye salmonscales. International Pacific Salmon Fisheries Commission Bulletin 9, 1–159.

Damsgard, B. & Dill, L. M. (1998). Risk-taking behaviour in weight compensating cohosalmon, Oncorhynchus kisutch. Behavioural Ecology 9, 26–32.

Einum, S., Thorstad, E. B. & Næsje, T. F. (2002). Growth rate correlations across life-stagesin female Atlantic salmon. Journal of Fish Biology 60, 780–784.

Farley, E. V. Jr., Murphy, J. M., Eisner, L., Helle, J. H., Moss, J. H., Adkison, M. D. &Nielsen, J. L. (2007). Critical-size distribution and size selective mortality of juvenileBristol Bay sockeye salmon. Fishery Bulletin 105, 121–130.

Farley, E. V. Jr., Murphy, J., Moss, J., Feldman, A. & Eisner, L. (2009). Marine ecologyof western Alaska juvenile salmon. In Pacific Salmon: Ecology and Management ofWestern Alaska’s Populations (Krueger, C. C. & Zimmerman, C. E., eds) pp. 307–329.American Fisheries Society, Symposium 70.

Fisher, J. P. & Pearcy, W. G. (2005). Seasonal changes in growth of coho salmon(Oncorhynchus kisutch) off Oregon and Washington and concurrent changes in thespacing of scale circuli. Fishery Bulletin 103, 34–51.

Friedland, K. D., Clarke, L. M., Dutil, J. D. & Salminen, M. (2006). The relationship betweensmolt and post-smolt growth for Atlantic salmon (Salmo salar) in the Gulf of St.Lawrence. Fishery Bulletin 104, 149–155.

Fukuwaka, M. & Kaeriyama, M. (1997). Scale analyses to estimate somatic growth in sockeyesalmon, Oncorhynchus nerka. Canadian Journal of Fisheries and Aquatic Sciences 54,631–636.

Good, T. P., Waples, R. S. & Adams, P. (2005). Updated status of federally listed ESUs ofWest Coast salmon and steelhead. U.S. Department of Commerce, NOAA TechnicalMemorandom, NMFS-NWFSC-66.

Gross, M. R. (1987). Evolution of diadromy in fishes. In Common Strategies of Anadro-mous and Catadromous Fishes (Dadswell, M. J., Kluada, R. J., Moffit, C. M., Sanders,R. L., Rulifson, R. A. & Cooper, J. E., eds), pp. 14–25. American Fisheries SocietySymposium 1.

Gross, M. R., Coleman, R. M. & McDowall, R. M. (1988). Aquatic productivity and the evo-lution of diadromous fish migration. Science 239, 1291–1293.



Hamon, T. R. & Foote, C. J. (2005). Concurrent natural and sexual selection in wild malesockeye salmon, Oncorhynchus nerka. Evolution 59, 1104–1118.

Hare, S. R., & Mantua, N. J. (2000). Empirical evidence for North Pacific regime shifts in1977 and 1989. Progress in Oceanography 47, 103–146.

Healey, M. C. (1982). Timing and relative intensity of size-selective mortality of juvenilechum salmon, Oncorhynchus keta, during early sea life. Canadian Journal of Fisheriesand Aquatic Sciences 39, 952–957.

Healey, M. C. (1986). Optimum size and age at maturity in Pacific salmon and effects of size-selective fisheries. Canadian Special Publication of Fisheries and Aquatic Sciences 89,39–52.

Healey, M. C. (1991). Life history of Chinook salmon. In Pacific Salmon Life Histories(Groot, C. & Margolis, L., eds), pp. 310–393. Vancouver, BC: University of BritishColumbia Press.

Helle, J. J. (1989). Relation between size-at-maturity and survival of progeny in chum salmon,Oncorhynchus keta (Walbaum). Journal of Fish Biology 35 (Suppl. A), 99–107.

Henderson, M. A. & Cass, A. J. (1991). Effect of smolt size on smolt-to-adult survival forChilko Lake sockeye salmon. Canadian Journal of Fisheries and Aquatic Sciences 48,988–994.

Hunt, G. L., Stabeno, P., Walters, G., Sinclair, E., Brodeur, R.D., Napp, J.M., & Bond, N.A.(2002). Climate change and control of the southeastern Bering Sea pelagic ecosystem.Deep Sea Research II 49, 5821–5853.

Johnsson, J. I., Blackburn, J., Clarke, W. C. & Withler, R. E. (1997). Does presmolt growthrate in steelhead trout (Oncorhynchus mykiss) and coho salmon (Oncorhynchus kisutch)predict growth rate in seawater? Canadian Journal of Fisheries and Aquatic Sciences54, 430–433.

Jonsson, N., Jonsson, B. & Fleming, I. A. (1996). Does early growth cause a phenotypicallyplastic response in egg production of Atlantic salmon? Functional Ecology 10, 89–96.

Juanes, F. (1994). What determines prey size selectivity in piscivorous fishes? In Theory andApplication in Fish Feeding Ecology (Stouder, D. J., Fresh, K. L. & Feller, R. J., eds),pp. 79–100. Columbia, SC: University of South Carolina Press.

Kaeriyama, M., Nakanura, M. Yamaguchi, M., Ueda, H., Anma, G., Takagi, S., Aydin, K. Y.,Walker, R. V. & Meyers, K. W. (2000). Feeding ecology of sockeye and pink salmonin the Gulf of Alaska. North Pacific Anadromous Fish Commission Bulletin 2, 55–63.

Kelley, E. R. (2001). Demographic responses to food and space competition by juvenilesteelhead trout. Ecology 82, 1247–1259.

Koenings, J. P., Geiger, H. J. & Hasbrouck, J. J. (1993). Smolt-to-adult survival patterns ofsockeye salmon: effects of smolt length and geographic latitude when entering the sea.Canadian Journal of Fisheries and Aquatic Sciences 49, 600–611.

Koo, T. S. (1962). Studies of Alaska Red Salmon . Seattle, WA: University of WashingtonPress.

Kruse, G. H. (1998). Salmon run failures in 1997–1998: a link to anomalous ocean condi-tions? Alaska Fishery Research Bulletin 5, 55–63.

Kutner, M. H., Nachtsheim, C. J., Neter, J. & Li, W. (2005). Applied Linear StatisticalModels . Boston, MA: McGraw-Hill.

Larsen, W. A. & McCleary, S. J. (1972). The use of partial residual plots in regression anal-ysis. Technometrics 14, 781–790.

Major, R. L., Ito, J., Ito, S. & Godfrey, H. (1978). Distribution and origin of Chinook salmon(Oncorhynchus tshawytscha) in offshore waters of the North Pacific Ocean. Interna-tional North Pacific Fisheries Commission Bulletin 38, 1–54.

Mortensen, D., Wertheimer, A., Taylor, S. & and Landingham, J. (2000). The relation betweenearly marine growth of pink salmon, Oncorhynchus gorbuscha, and marine temperature,secondary production and survival to adulthood. Fishery Bulletin 98, 319–335.

Moss, J. H., Beauchamp, D. A., Cross, A. D., Myers, K., Farley, E. V. Jr., Murphy, J. M.& Helle, J. H. (2005). Higher marine survival associated with faster growth for pinksalmon (Oncorhynchus gorbuscha). Transactions of the American Fisheries Society134, 1313–1322.



Nicieza, A.G. & Brana, F. (1993). Relationships among smolt size, marine growth, and seaage at maturity of Atlantic salmon (Salmo salar) in northern Spain. Canadian Journalof Fisheries and Aquatic Sciences 50, 1632–1640.

Peyronnet, A., Friedland, K. D., Maoileidigh, N. O., Manning, M., & Poole, W. R. (2007).Links between patterns of marine growth and survival of Atlantic salmon Salmosalar, L. Journal of Fish Biology 71, 684–700.

Price, C. S. & Schreck, C. B. (2003). Stress and seawater-entry behavior of juvenile Chinooksalmon (Oncorhynchus tshawytscha): conflicts in physiological motivation. CanadianJournal of Fisheries and Aquatic Sciences 60, 910–918.

Quinn, T. P. (2005). The Behavior and Ecology of Pacific Salmon and Trout . Seattle, WA:University of Washington Press.

Quinn, T. P. & Myers, K. W. (2005). Anadromy and the marine migrations of Pacific salmonand trout: Rounsefell revisited. Reviews in Fish Biology and Fisheries 14, 421–442.

Quinn, T. P., Vøllestad, L. A., Peterson, J. & Gallucci, V. (2004). Influences of freshwaterand marine growth on the egg size-egg number tradeoff in coho and Chinook salmon.Transactions of the American Fisheries Society 133, 55–65.

Ruggerone, G. T. (1992). Threespine stickleback aggregations create potential predationrefuge for sockeye salmon fry. Canadian Journal of Zoology 70, 1052–1056.

Ruggerone, G. T. & Goetz, F. A. (2004). Survival of Puget Sound Chinook salmon(Oncorhynchus tshawytscha) in response to climate-induced competition with pinksalmon (Oncorhynchus gorbuscha). Canadian Journal of Fisheries and Aquatic Sci-ences 61, 1756–1770.

Ruggerone, G. T., Zimmermann, M., Meyers, K. W., Nielsen, J. L. & Rogers, D. E. (2003).Competition between Asian pink salmon and North American sockeye salmon in theNorth Pacific Ocean. Fisheries and Oceanography 12, 209–219.

Ruggerone, G. T., Farley, E., Nielsen, J. & Hagen, P. (2005). Seasonal marine growth ofBristol Bay sockeye salmon (Oncorhynchus nerka) in relation to competition withAsian pink salmon (O. gorbuscha) and the 1977 ocean regime shift. Fishery Bulletin103, 355–370.

Ruggerone, G. T., Nielsen, J. L. & Bumgarner, J. (2007). Linkages between Alaskan sockeyesalmon abundance, growth at sea, and climate, 1955–2002. Deep Sea Research II 54,2776–2793.

Salminen, M. (1997). Relationships between smolt size, post-smolt growth and sea age atmaturity in Atlantic salmon ranched in the Baltic Sea. Journal of Applied Ichthyology13, 121–130.

Salminen, M. (2002). Marine survival of Atlantic salmon in the Baltic Sea. North PacificAnadromous Fish Commission Technical Report 4, 27–29.

Skilbrei, O. T. (1989). Relationships between smolt length and growth and maturation in thesea of individually tagged Atlantic salmon (Salmo salar). Aquaculture 83, 95–108.

Snover, M. L., Watters, G. M. & Mangel, M. (2005). Interacting effects of behavior andoceanography on growth in salmonids with examples for coho salmon (Oncorhynchuskisutch). Canadian Journal of Fisheries and Aquatic Sciences 62, 1219–1230.

Vøllestad, L. A., Peterson, J. & Quinn, T. P. (2004). Effects of freshwater and marine growthrates on early maturation in male coho and Chinook salmon. Transactions of the Amer-ican Fisheries Society 133, 495–503.

Whitmore, C., Martz, M., Linderman, J. C. Jr., Fisher, R. L. & Bue, D. G. (2008). Annualmanagement report for the subsistence and commercial fisheries of the Kuskokwimarea, 2004. Fisheries Management Report No. 08-25. Anchorage, AK: Alaska Depart-ment of Fish and Game.

Yoshiyama, R. M., Fisher, F. W. & Moyle, P. B. (1998). Historical abundance and declineof Chinook salmon in the Central Valley region of California. North American Journalof Fisheries Management 18, 487–521.

Electronic Reference

Hagen, P. T., Oxman, D. S. & Agler, B. A. (2001). Developing and Deploying a High Res-olution Imaging Approach for Scale Analysis. Document 567. Vancouver, BC: NorthPacific Anadromous Fisheries Commission. Available at www.npafc.org


linking marine and freshwater growth in … marine and freshwater growth in western alaska chinook...

Documents