John Carlos GarzaEric C. AndersonTommy Williams

Brian SpenceKerrie Pipal

Libby Gilbert-Horvath

John Carlos GarzaEric C. AndersonTommy Williams

Brian SpenceKerrie Pipal

Libby Gilbert-Horvath

Spatial and temporal variation in structure and abundance of coho

salmon in California

Status of coho salmon in California, USA

• Population declines, local extinctions

• Management units: Evolutionarily Significant Units (ESUs)

– Southern Oregon/Northern California Coast Coho Salmon ESU (SONCC) –Threatened

– Central California Coast Coho Salmon ESU (CCC) – Endangered

• All populations have protected status

• Life history– Anadromous, reproduce in natal streams,

semelparous

– 3-year life cycle; 3 brood years, little temporal gene flow

SONCC

CCC

Oregon

California

Gilbert-Horvath et al. (in press), 18 microsatellite loci

Background: phylogeography of coho salmon in California

• 30 sites surveyed for young-of-year in 2003

• Significant hierarchical genetic structure– Greatest divergence at

broadest geographic scale

• Klamath River basin distinct from other northern populations

Central California ESUNorthern California ESUKlamath River basin

Gilbert-Horvath et al. (in press), 18 microsatellite loci

Background: phylogeography of coho salmon in California

Bayesian Clustering Analysis-Individual Genotypes

Gilbert-Horvath et al. (in press), 18 microsatellite loci

Background: phylogeography of coho salmon in California

Klamath South Coast (Ten Mile, Noyo, Lagunitas, Scott, etc)

North Coast (Mad, Eel, Mattole, etc)

Factorial Correspondence Analysis-Individual Genotypes

Study design: temporal monitoring• Survey populations at two time points

– Time series data at a regional scale– Examine allele frequency change over time

• Timepoint 1: baseline sampling in rivers throughout California in 2003 (n=1,976)

• Timepoint 2: revisited same sites in 2015 (n=2,223)– Expanded to include Oregon sites

• Four generations between surveys– Sampled two cohorts of the same brood cycle

Sampling locations• Sampled juveniles in

natal streams• Spatially stratified

electrofishing protocol• 46 sites yielded coho

salmon in 2015• n=75 maximum per site California

2003 survey2015 survey

A. McHuron

43o N

37o N

Oregon

Analytical objectives• Analysis of family structure• Examine patterns of population

structure at multiple spatial scales• Assess temporal stability of genetic

composition of populations • Estimate effective population size (Ne)

Genotype data: 95 SNP locin=4,199 individuals

A. McHuron

Results• Sibship reconstruction: full

siblings detected in almost all populations

• Mean sibship size by site -> • Mean sibship size by year

• 2003 = 2.2• 2015 = 3.3

• Family structure may obscure or distort population structure– Omit full sibs from population

analyses (revised n=2,460)

20032015

=

Results from Colony v2.0.6.1 (Wang 2008, Jones and Wang 2010)

0

20

40

60

80

100

0 20 40 60 80 100

Num

ber o

f spa

wer

s pe

r pop

ulat

ion,

201

5

Number of spawners per population, 2003

Temporal trends in reproduction• 32 paired temporal

population samples

• Estimated number of spawning adults that produced each sample of juveniles

• Net decrease in number of spawners between 2003 and 2015

Results from Colony v2.0.6.1

C. Nicol

Paired population sample

Effective population size• Method in development: estimating Ne using the temporal method,

accounting for family structure using Colony output• Upper bound of Ne difficult to estimate due to small sample sizes

Population

central California northern California Klamath River (northern)

0

50

100

150

200

250

300

350

400

WAG RED KT

D

CAS NVJ

KSH

TNL

LIT

NO

Y

ALB

KCC

LOL

SCC

KSS

BSA

BIG

LDG

LAG

PUD

RPR

FWC

EHT

KMG

KTS

LTU

Ne

(Max

imum

Lik

elih

ood

Estim

ate) MLE lower bound of Ne

Lower Confidence Interval

Upper Confidence Interval

Temporal trends in reproduction, by population: estimated number of spawning adults in 2003 and 2015

0

10

20

30

40

50

60

70

Cou

nt

Population samples (north to south)

Winter 2003

Winter 2015

0

0.1

0.2

0.3

0.4

0.5

0.6

-0.05 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

Rel

ativ

e pr

opor

tion

of p

airw

ise

com

paris

ons

FST

Between ESUs (n=781 comparisons)

Basins within SONCC (n=354)

Basins within CCC (n=322)

Tributaries within basin (n=77)

Population between years (n=30)

Relative divergence between populations: distributions of pairwise

FST estimates

100

ROAb

WILb

WILa

75

82

TNCb

TNRb

50

TNLa

TNLb NOYb

74

PUDb

PUDa

COTaLOLb

LOLa LDGa

97

REDb REDa

WLKbLAGa

61

LAGbLDGb

ALBb

LITbLITa

91

GCSb

GCNb

NVJbNVFb

ALBaNVJa

BIGaBIGb

BSAaBSAb

99 SCCa

SCCbCASb

CASaNOYa

WAGaWAGb

85

MTTa

MTTbMDSa

MDSb86

FWCb

FWCa

99

EHTb

EHTa

52

82

LTUa

LTUb

MPLbRPRbMPLa

RPRa

96

CHEbCHJb

KMGa

KBVaROVb

ROKbKSHaKSCa

KSCb

KSSaKSSb

53

58

KTSb 85

KTDb

KTDa

KTSa

ROTb

KSHb

KCCa

ROIb

KCCb

KMGb

Phylogeographic structure

• High-level topology concordant with geography and management units

• Tributaries within basins cluster together, except in largest basins (Klamath & Rogue rivers)

• Temporally coherent genetic composition in majority of paired temporal samples

Southern Oregon/Northern

Calif. ESU

Central Calif. ESU

N

Full siblings omitted

100

ROAb

WILb

WILa

75

82

TNCb

TNRb

50

TNLa

TNLb NOYb

74

PUDb

PUDa

COTaLOLb

LOLa LDGa

97

REDb REDa

WLKbLAGa

61

LAGbLDGb

ALBb

LITbLITa

91

GCSb

GCNb

NVJbNVFb

ALBaNVJa

BIGaBIGb

BSAaBSAb

99 SCCa

SCCbCASb

CASaNOYa

WAGaWAGb

85

MTTa

MTTbMDSa

MDSb86

FWCb

FWCa

99

EHTb

EHTa

52

82

LTUa

LTUb

MPLbRPRbMPLa

RPRa

96

CHEbCHJb

KMGa

KBVaROVb

ROKbKSHaKSCa

KSCb

KSSaKSSb

53

58

KTSb 85

KTDb

KTDa

KTSa

ROTb

KSHb

KCCa

ROIb

KCCb

KMGb

Phylogeographic structure

• High-level topology concordant with geography and management units

• Tributaries within basins cluster together, except in largest basins (Klamath & Rogue rivers)

• Temporally coherent genetic composition in majority of paired temporal samples

NTributary samples

100

ROAb

WILb

WILa

75

82

TNCb

TNRb

50

TNLa

TNLb NOYb

74

PUDb

PUDa

COTaLOLb

LOLa LDGa

97

REDb REDa

WLKbLAGa

61

LAGbLDGb

ALBb

LITbLITa

91

GCSb

GCNb

NVJbNVFb

ALBaNVJa

BIGaBIGb

BSAaBSAb

99 SCCa

SCCbCASb

CASaNOYa

WAGaWAGb

85

MTTa

MTTbMDSa

MDSb86

FWCb

FWCa

99

EHTb

EHTa

52

82

LTUa

LTUb

MPLbRPRbMPLa

RPRa

96

CHEbCHJb

KMGa

KBVaROVb

ROKbKSHaKSCa

KSCb

KSSaKSSb

53

58

KTSb 85

KTDb

KTDa

KTSa

ROTb

KSHb

KCCa

ROIb

KCCb

KMGb

Paired temporal population samples

Phylogeographic structure

• High-level topology concordant with geography and management units

• Tributaries within basins cluster together, except in largest basins (Klamath & Rogue rivers)

• Temporally coherent genetic composition in majority of paired temporal samples

N

Fractional ancestryW

ILa

WIL

bKC

CaKC

CbKM

GaKM

GbKS

CaKS

CbKS

HaKS

HbKS

SaKS

Sb

KTSa

KTSb

KTDa

KTDb

RPRa

RPRb

MPL

aM

PLb

LTUa

LTUb

MDS

aM

DSb

FWCa

FWCb

EHTa

EHTb

MTT

aM

TTb

WAG

aW

AGb

TNLa

TNLb

PUDa

PUDb

NOYa

NOYb

CASa

CASb

BIGa

BIGb

LITa

LITb

ALBa

ALBb

BSAa

BSAb

NVJa

NVJb

LAGa

LAGb

LOLa

LOLb

LDGa

LDGb

REDa

REDb

SCCa

SCCb

K = 2

K = 3

K = 4

K = 5

3/5

K = 6

1/5

1/5

5/5

4/5

1/5

5/5

5/5

Structure v2.3.4 on paired temporal populations, full siblings omitted

Genetic analysis of Mattole River cohosalmon young-of-year

• Baker Creek tributary, Mattole River, California

• YOY (young-of-year) were sampled – August 2013 (Brood year 12/13)– August 2014 (Brood year 13/14)

• Genetic analyses – 90 SNP loci– Colony v2 to resolve familial relationships– Population assignment test using nearby

reference populations– Ancestry analysis

Ancestry analysis (Structure v2.3.4)CO

Ta

EHTa

EHTb

FWCa

FWCb

MDS

a

MDS

b

MTR

MTT

a

MTT

b

MTT

Ba

MTT

Bb

MTT

C

PUDa

PUDb

TNLa

TNLb

K = 2 (5/5)

K = 3 (3/5)

K = 3 (2/5)

K = 4 (3/5)K = 4 (2/5)

K = 5 (4/5)

K = 5 (1/5)

K = 6 (2/5)

K = 6 (1/5)

K = 6 (1/5)

K = 6 (1/5)

Cot

tane

va

Eel RiverFreshwater Ck Mad R

Mattole R Ten Mile RPudding Ck

MTTBa: BY12MTTBb: BY13

Conclusions• Statewide decrease in number of spawning adults that

produced the sampled juveniles– Impact of 5-year drought on habitat

• Regional-scale geographic structure remains concordant with management unit boundary

• Significant spatial population structure – Particularly between ESUs, and among populations of

northern ESU • Little divergence between tributaries of the same basin• Minimal temporal structure: within-population genetic

composition relatively consistent over time• Ancestry and affiliation of small populations influenced

by migrants from nearby basins

Implications for conservation• Connectivity of populations• Extinction risk in susceptible small populations

– Benefit of straying from productive populations• Periodic re-evaluation of strategies

– Identify populations or areas of concern– Where are recovery efforts having a positive impact?

A. McHuron

AcknowledgementsAcknowledgements• NOAA Fisheries Service, Southwest Fisheries Science

Center, Fisheries Ecology Division– Molecular Ecology Team: Ellen Campbell, Anthony Clemento, Cassie

Columbus, Laney Correa – Landscape Ecology Team: Heidi Fish, Alex McHuron, Casey

Sheridan, Emily Tucker

• California Department of Fish & Wildlife

Funding Sources

National Marine Fisheries Service

Fisheries Restoration Grant from the California Department of Fish & Wildlife M. Bond

Photo Copyright (C) 2002-2015 Kenneth & Gabrielle Adelman, California Coastal Records Project, www.Californiacoastline.org

Thank you!