Factors affecting the marine survival of Puget Sound steelhead

Northwest Fisheries Science Center

Barry Berejikian and Megan MooreEnvironmental and Fisheries Sciences Division

Manchester Research Station

Salish Sea Marine Survival Project(Steelhead Workgroup)

Puget Sound Steelhead Marine Survival Workgroup Participants

• Neala Kendall, WDFW• Megan Moore, NWFSC• Barry Berejikian, NWFSC• Scott Pearson, WDFW• Ken Warheit, WDFW• Erik Neatherlin, WDFW• Chris Ellings, Nisqually Indian Tribe*• Sandie O’Neill, WDFW• Mike Crewson, Tulalip Tribes*• Ed Connor, Seattle City Light

U.S. Department of Commerce | National Oceanic and Atmospheric Administration | NOAA Fisheries | Page 2

Contributing Experts• Steve Jeffries, WDFW• Bruce Stewart, NWIFC• Paul Hershberger, USGS• John Kerwin, WDFW• Dave Beauchamp, UW• Linda Rhodes, NWFSC• Lyndal Johnson, NWFSC• Gina Yitalo, NWFSC• Penny Swanson, NWFSC• Brian Beckman, NWFSC• Andy Goodwin, U.S. Fish and Wildlife Service• Joy Evered, U.S. Fish and Wildlife Service• Kym Jacobson, NWFSC• Mary Arkoosh, NWFSC• Joe Dietrich, NWFSC

Project Management and Facilitation• Michael Schmidt, Long Live the Kings• Iris Kemp, Long Live the Kings

3

Outline

• Puget Sound steelhead life history and abundance trends

• Factors that may affect marine survival

• Freshwater

• Marine

• Salish Sea Marine Survival Project (wrt steelhead) – next steps

Puget Sound Steelhead

• Very popular sport and important tribal fisheries

• Predominantly winter-run populations (late fall – spring)

• Roughly 1.5 M hatchery-reared fish released annually

• PS steelhead listed ‘Threatened’ in 2007

• Factors in listing decision

• Declines in abundance and productivity

• Habitat (dams, urbanization, water quality)

• Artificial propagation

Oncorhynchus mykiss life history

Egg

Parr

Maturation in freshwater

Smolt (age 1 -3)

Estuary/early marine (2 weeks)

Ocean (1 – 3 years)

Maturation and spawning Rainbow trout

(resident form)

Steelhead(anadromous form)

Abundance of Puget Sound and WA coast populations

1980 1985 1990 1995 2000 2005 2010 20150

5000

10000

15000

20000

25000

SkagitGreenPuyallupNisquallyQueetsQuinaultHohQuil-layute

Abundance trends

Nisqually Puyallup Green Skagit Quinault Queets Hoh Quil-layute

0%

10%

20%

30%

40%

50%

60%

70%

80%Puget Sound Washington Coast

SouthNorth

SouthNorth

Mea

n (2

005-

2010

)/Mea

n (1

984-

1989

)

Marine (smolt-to-adult) survival

0%

2%

4%

6%

8%

10%Lower Columbia Coast Puget Sound

SAR

1970 1975 1980 1985 1990 1995 2000 2005 20100%

2%

4%

6%

8%

10%

Brood year

SAR

Summer-run

Winter-run

Source: Puget Sound Steelhead Marine Survival Draft Workplan

Where does this lead us?….• Abundance and SAR trends point to:

Strong marine signal Different signal within Puget Sound than elsewhere (lower survival), particularly since

early1990’s Possibly worse conditions in ‘the deep south’

• What is different about Puget Sound that might reduce marine survival relative to other regions?

Puget Sound freshwater stream effects on smolt characteristics

Migration routes in the Pacific Ocean

Puget Sound marine conditions

Freshwater effects (on subsequent marine survival)

• Some hypotheses:Reduced diversity (‘Portfolio effect’: e.g., Schindler et al. 2012. Nature) Hatchery effects (genetic or ecological)Water quality (e.g., toxics)Disease

11

Hood Canal Steelhead

Freshwater: Spawn timing

January February March April May June July0.000.100.200.300.400.500.600.700.800.90

Dewa

tto R

...

Unio

n Ri

ver

SF S

koko

m...

Duck

abus

h

Littl

e Q

uil..

.

Prop

ortio

n of

redd

s

Source: WDFW SGS Database & Hood Canal Steelhead ProjectJa

nuary

February

March

April

MayJu

ne July

0.000.050.100.150.200.250.300.350.400.450.50

Kitsap PeninsulaOlympic Peninsula

Prop

ortio

n of

redd

s

13

Freshwater: Spawn timing and size at age

50 60 70 80 90 100 110 120 130 140 15080

90

100

110

120

130

140

R² = 0.883604525182642

Mean spawning date (days from Jan 1)

Mea

n fo

rk le

ngth

(mm

) at a

ge-1

(spr

ing)

Age-1 Age-2 Age-30

20406080

100120140160180200

Fork

leng

th (m

m)

Source: Hood Canal Steelhead Project 14

Big Bee

f

Big Bee

f

Big Bee

f

Big Bee

f

Dewatt

o

Dewatt

o

Dewatt

o

Dewatt

o

Tahuy

a

Tahuy

a

Tahuy

a

Tahuy

a

Little

Quilc

ene

Little

Quilc

ene

Little

Quilc

ene

Little

Quilc

ene

Ducka

bush

Ducka

bush

Ducka

bush

Ducka

bush

Hamma H

amma

Hamma H

amma

Hamma H

amma

Hamma H

amma

SF Skoko

mish

SF Skoko

mish

SF Skoko

mish

SF Skoko

mish0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1 2 3 4

Age

Freshwater: Age-at-smoltification

Source: Hood Canal Steelhead Project

20-Mar 3-Apr 17-Apr 1-May 15-May 29-May 12-Jun 26-Jun0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1 Big BeefDewattoTahuyaLittle QuilceneDuckabushHamma HammaSkokomish

Date

Cum

ulat

ive

prop

ortio

n of

sm

olts

Freshwater: Smolt migration timing

360 410 460 510 560 610105

110

115

120

125

130

135

L. Quilcene

Duckabush

Hamma Hamma SF Skokomish

TahuyaDewatto

Big Beef

R² = 0.645515286222524

Accumulated winter temperature units ( C) M

ean

Cale

ndar

day

of o

utm

igra

tion

16

Other freshwater effectsHatcheries may have both ecological and genetic effects: Notably, no steelhead released into the Nisqually River since 1980

Water Quality/toxics: Toxic contaminant exposure data for steelhead is lackingNisqually considered most pristine in main basin of Puget Sound

Disease: Nanophyetus: more prevalent in south than north Puget Sound Rapid infections; can affect swimming performance

Nisqually Puyallup Green Skagit Quinault Queets Hoh Quillayute0%

10%

20%

30%

40%

50%

60%

70%

80%

Mea

n(20

05-2

010)

/Mea

n(19

84-1

989)

Puget Sound Washington Coast

Mean annual hatchery smolt releases 1986-2007

0K 203K 207K 329K

Steelhead in the marine environment

• Pacific Ocean migratory patterns and distribution

• Puget Sound migratory behavior and survival

Steelhead Ocean migratory behavior

McMichael et al. Animal Biotelemetry 2013, 1:14 (Figures 2 and 4)

19

Steelhead Ocean Distribution (spring/summer)

Atcheson et al. 2010. Trans. Am. Fish. Soc. 141 (4)

Steelhead in the marine environment

• Pacific Ocean migratory patterns and distribution

• Puget Sound migratory behavior and survival

21

TABLE 2. Fish taxa captured by surface trawls at 52 sites in greater Puget Sound during May–August 2003; taxa are ranked in order based on highest to lowest frequency of occurrence. (Rice et al. 2012. Marine and Coastal Fisheries 4: 117-128)

SPECIES% frequency

1. Chinook salmon Oncorhynchus tshawytscha 65.62. Pacific herring Clupea pallasii

57.63. Threespine stickleback Gasterosteus aculeatus 51.54. Surf smelt Hypomesus pretiosus

50.05. Chum salmon O. keta

35.46. River lamprey Lampetra ayresii

25.47. Pacific sand lance Ammodytes hexapterus 22.48. Coho salmon O. kisutch

11.79. Bay pipefish Syngnathus leptorhynchus 11.210. Pacific sandfish Trichodon trichodon

9.011. Starry flounder Platichthys stellatus

8.812. Shiner perch Cymatogaster aggregata

6.113. Steelhead O. mykiss

3.7

**species ranked 14-33 not shown

Acoustic telemetry

Vemco 7mm and 9 mm pingers@ 69kHz, 136 db

Limitations/considerations for acoustic telemetry

• Handling and tag effects never fully known (some negative effects on growth)

• Tag loss in seawater 2% (V7) - 12% (V9), but not until after outmigration

• Seals can hear them (Cunningham et al. in review)

• Detection range of receivers varies depending on currents, noise, water quality etc.

• Expected to under-estimate natural survival rates

Steelhead telemetry in Hood Canal

Moore et al. 2010a. TAFSMoore et al. 2010b. PLOS OneMoore et al. 2012. PLOS OneMoore et al. 2013. PLOS One

25Moore et al. 2010a. Trans. Am. Fish. Soc.

Moore M, Berejikian BA, Tezak EP (2013) A Floating Bridge Disrupts Seaward Migration and Increases Mortality of Steelhead Smolts in Hood Canal, Washington State. PLoS ONE 8(9): e73427. doi:10.1371/journal.pone.0073427http://www.plosone.org/article/info:doi/10.1371/journal.pone.0073427

Known survivors

Presumed mortalities

Telemetry useful for identifying hotspots

Possible mortalities

Undetermined

Puget Sound Telemetry ‘Study’

Hood Canal Rivers: 2006-2010Moore, Berejikian, et al. (NWFSC)

Green River: 2006-2009Fred Goetz, Tom Quinn et al (UW, ACOE)

Puyallup River: 2006, 2008-2009Andrew Berger et al. (Puyallup Tribe)

Nisqually River: 2006-2009Sayre Hodgson et al. (Nisqually Tribe)

Skagit River: 2006-2009Ed Conner et al. (Seattle City Light)

Telemetry array

Migration SegmentsHood Canal Puget Sound Skagit

1st segment River Mouth - HCB River Mouth - CPS River Mouth - DP2nd segment HCB – ADM CPS – ADM 3rd segment ADM – JDF ADM – JDF DP - JDF

2006 -2009 Puget Sound Telemetry Study

Population

N2006 N2007 N2008 N2009

W H W H W H W H

Hood canal 73 33 123 47 67 42 105 59

Green 100 50 39 50 48 50 50 -

Nisqually 55 - 49 - 14 - 69 -

Puyallup 25 25 - - - 90 - 66

Skagit 23 - 47 - 50 50 25 55

TOTAL 334 355 411 293

Cumulative travel times

RM-HCB RM-ADM RM-JDF0

2

4

6

8

10

12

14

16

18

20Hood CanalSkok

BBC

RM-NAR RM-CPS/DP RM-ADM RM-JDF0

2

4

6

8

10

12

14

16

18

20Puget Sound

NisquallyPuyallupGreenSkagit

Trav

el ti

me

(day

s)

Travel rates

RM-NAR/CPS/DP

NAR-CPS CPS-ADM ADM/DP-JDF0

5

10

15

20

25

30

35

40

45

50Puget SoundNisqually

PuyallupGreenSkagit

Trav

el ra

te (k

m/d

)

RM-HCB HCB-ADM ADM-JDF0

5

10

15

20

25

30

35

40

45

50Hood Canal

Skok

BBC

Mark-Recapture Survival Estimates: Cormack-Jolly-Seber

Population

N2006 N2007 N2008 N2009

W H W H W H W H

Hood canal 73 33 123 47 67 42 105 59Green 100 50 39 50 48 50 50 -Nisqually 55 - 49 - 14 - 69 -Puyallup 25 25 - - - 90 - 66Skagit 23 - 47 - 50 50 25 55

TOTAL 334 355 411 293

Categorical variablesPopulation

Region (HC, SS, Skagit)Rear type (H/W)Migration SegmentYearTag Type

Continuous variablesDistance travelledBody Length

Model with lowest AICc = ~(Segment x population)+ (year)+ (rear type H/W)

N=1393

Early marine survival in Hood Canal and Puget Sound steelhead

Skokomish (W)

Skokomish (H) Hamma (H) BBC (W)0%

5%

10%

15%

20%

25%

30%

35%

40%

Hood Canal

Aver

age

RM-J

DF S

urvi

val

Nisqually (W)

Puyallup (W)

Puyallup (H) Green (W) Green (H) Skagit (W) Skagit (H)0%

5%

10%

15%

20%

25%

30%

35%

40%Puget Sound

W = Wild H = Hatchery

Combined early marine survival estimate = 17% (hatchery = 12% , wild = 20%)

Source: Welch et al. PNAS 108: 8708-8713

Strait of Georgia and Johnstone Strait steelhead survival estimates

35

1 37 73 1091451812172532893253613974334695055415776136496857210

1000

2000

3000

4000

5000

6000

Days after seawater entry

Num

ber o

f sur

vivo

rs

River mouth to Pacific Ocean (21 days)

Adult return at 3% SAR (720 days)

Instantaneous daily mortality rates: River mouth to Pacific Ocean (M = 0.086)Pacific Ocean entry to adult return (M = 0.003)

Distance from Juan de Fuca Strait (km)

050100150200250

Sur

viva

l (%

)

0

20

40

60

80

100

Big Beef Green Nisqually Puyallup Skagit Skokomish

Can we identify mortality ‘hot spots’?

NAR

CPS

HCB

DP

ADM

NAR = Tacoma NarrowsCPS = Central Puget SoundADM = Admiralty InletHCB = Hood Canal BridgeDP = Deception Pass

Comparing Puget Sound to Hood Canal

RM-HCB/CPS HCB/CPS-ADM ADM-JDF RM-JDF0%

10%20%30%40%50%60%70%80%90%

100%

2008Su

rviva

l Pro

babi

lity

RM-HCB/CPS HCB/CPS-ADM ADM-JDF RM-JDF0%

10%20%30%40%50%60%70%80%90%

100% 2009

(Skokomish RM-HCB = 66 km)(Nisqually RM – CPS = 67 km)

Telemetry Summary Low early marine survival rates consistent with SARs and abundance of Puget Sound steelhead

Instantaneous daily mortality rates are high (i.e., mortality occurs very quickly)

Rapid travel times (1 – 3 weeks from river mouths to ocean entry)

Puget sound may serve more a migratory corridor than a rearing environment

Has it always been this way, or is that part of the problem?

Hood Canal and Puget Sound steelhead exhibit different patterns

Central Puget Sound may represent a mortality hotspot

Steelhead dying at high rate in PS

Predation IS proximate/ direct cause of mortality

Predation IS NOT proximate/ direct cause of mortality

Poor fish condition and/or altered behavior:

freshwater (F) or marine (M) effect (ranked)

1. Disease (M/F) 2. Poor water quality/toxics (M/F)3. Genetic fitness loss (F)4. HABs (M)5. Foraging/Starvation (M)6. Portfolio (outmigrant size or timing -F)7. Structural changes in marine habitats (M)

Steelhead Research Planning: thought process

Predator-prey interactions

1. Depensation2. Increase in key predator populations3. Decrease abundance of other prey

2013-2015 Steelhead Marine Survival Study

Activities2013

September

October

November

December

2014

January

February

March

April

May

June

July

August

September

October

November

December

2015

January

February

March

April

May

June

Permitting Upgrade telemetry receivers Studies

1: Complete retro telemetry data analysis

2: Complete SAR trend analysis 3: Fish characteristics vs. SARs 4. Enviro. data vs. SARs & telemetry 5: Predator review 6: Genome-wide association study 7: Juvenile fish health assessment 8: Reciprocal transplant 9. Harbor seal interactions 10: Dinner bell effect 11: Modeling (affiliated)

Legend for study work

Preparation Field Work Analysis Reporting Tech Memo Due

41

8. Reciprocal transplant

Test following effects on marine survival Population Location 1st segment 2nd segment Translocation (nuisance)

B1

C

B2

D

E

A2

A1

42

Predator-prey interactions

Avian predators (S. Pearson, WDFW, review in prep)cormorants (most abundant)Caspian ternscommon mergansersloons rhinoceros auklets (feed on smaller prey)

Mammalian predatorsHarbor porpoiseHarbor seals

Identifying potentially important predators on steelhead smolts

Criteria: 1. Spatial and temporal overlap, 2. Known to eat steelhead 3. Known to eat similarly sized salmon or other fish4. Increasing or stable abundance,

43

Predator-prey interactions (harbor porpoise)

(J. Evenson, WDFW, 2013, unpublished data)

1993-1998 1999-2004 2005-2011

44

Predator-prey interactions (harbor seals)Harbor seal counts

Jeffries et al. 2003 J. Wildlife Manage.

45

9. Harbor seal/steelhead interactions

B1

C

B2

D

E

A2

A1

Quantify

core foraging areas of harbor seals during the steelhead smolt outmigration;

spatial and temporal overlap of harbor seals and steelhead smolts in specific areas of Puget Sound;

predation events by seals on tagged steelhead smolts

GPS, acoustic tag/receiver, VHF

10. Dinner Bell effect

Do smolts with pinging transmitters experience similar mortality to those with temporarily silent transmitters?

10. Dinner Bell effect

Do smolts with pinging transmitters experience similar mortality to those with temporarily silent transmitters?

~ 5 days

More Acknowledgements…..

Telemetry study funding provided through NOAA, USACE (Seattle District), UW, Steelhead Trout Club of Washington, Pacific Ocean Shelf Tracking Network (POST)

Survival Modeling SupportMike Melnychuk (UW)Jeff Laake (NOAA SWFSC)

Field/Logistic SupportLong Live the Kings ▪ Hood Canal Salmon Enhancement Group ▪ Mat Gillam ▪ R2 Resource Consultants ▪ Bob Leland ▪ Kelly Kiyohara ▪ Pat Michael Brody Antipa ▪ Pete Topping ▪ Deborah Feldman ▪ Kelly Andrews ▪ John Blaine ▪ Jim Deveraux ▪ Skip Tezak Correigh Greene ▪ Anna Kagley ▪ Shawn Larson ▪ Jeff Christiansen ▪ John Rupp ▪ Chuck Ebel ▪ Jose Reyes-Tomassini ▪ Jennifer Scheurell ▪

Chris Ewing Dawn Pucci ▪ Kurt Dobszinsky ▪ Paul Winchell ▪ David Welch ▪ Debbie Goetz ▪ JoseGimenez ▪ Aswea Porter ▪ Emiliano Perez ▪ Craig Smith ▪ Tim Wilson ▪ Florian Leischner ▪ Christopher Ellings ▪ Scott Steltzner

• Figure 2. Survival estimates for smolts migrating through fresh- and saltwater migration segments.

Moore M, Berejikian BA, Tezak EP (2012) Variation in the Early Marine Survival and Behavior of Natural and Hatchery-Reared Hood Canal Steelhead. PLoS One 7(11): e49645. doi:10.1371/journal.pone.0049645http://www.plosone.org/article/info:doi/10.1371/journal.pone.0049645

Telemetry useful for estimating survival (especially relative survival)

-Release to river mouth-River mouth to HC bridge-HC bridge to Admiralty Inlet-Admiralty Inlet to Pillar Point

Comparing Puget Sound to Hood Canal

Surv

ival P

roba

bility

RM-HCB/CPS HCB/CPS-ADM ADM-JDF RM-JDF0%

10%20%30%40%50%60%70%80%90%

100%

2008

RM-HCB/CPS HCB/CPS-ADM ADM-JDF RM-JDF0%

10%20%30%40%50%60%70%80%90%

100%2009

(Big Beef RM-HCB = 25 km)(Green River RM-CPS = 11 km)

Quantifying encounter and predation rates

• Spatial overlap: seal and steelhead detections on same receivers

• Temporal and spatial overlap: concurrent pings on same fixed (moored) receivers

• Potential encounters: pings detected on mobile (seal-mounted) receivers

• Putative predation events: • recurring continuous pings on seal-mounted receivers,

perhaps followed by...• Stationary tags (defacated tags) connected to seal

movements/locations

Puget Sound Chinook salmon

• Residency in Puget Sound = weeks to years.• Very abundant in Puget Sound spring through

summer (Beauchamp and Duffy 2011, Rice et al. 2012)

• Spring/summer growth rate and body size strongly correlated with survival (in hatchery stocks: Duffy and Beauchamp 2011)

• Foraging opportunities, diet composition, and competition (including comp. with hatchery Chinook salmon) likely influence survival.

5304/10/2023

20 fish were released and 6 were detected at the Point Reyes Array. That’s 30% survival over 137 km

12 (60%) were detected traveling from the river mouth through San Pablo Bay and through San Francisco Bay (77 Km), 2 went south of the Bay Bridge Array, and 10 that entered the ocean were detected at point reyes.Sundstrom et al 2013

Oregon estuaries, lose about 50% in < 20 km or even shorter distances. Steepest losses in areas where there’s been documented high predation rates

San Francisco Bay

04/10/2023 54

Species Year Release date

Release size (mm)

Duration(weeks)

Detected (Chinook)Survival est (Sthd)

Instantaneous daily mortality

Chinook (Hatch)

2008 May 9 190-233 ~14 43% M = 0.008

Sthd (Wild)

2008 April 16 – May 27

170-190 ~ 2~ 1~ 3

89% RM to HCB18% HCB to JDF16.5% RM to JDF

M = 0.008M = 0.242M = 0.086

Is the estimate 20% EMS high or low?

Chamberlain et al. 2011 TAFS and Moore et al. 2012 PLOS One