Observing halibut survival after trawler deck-release with - - PowerPoint PPT Presentation

Observing halibut survival after trawler deck-release with satellite tags

Craig S. Rose, FishNext Research Julie Nielsen, Kingfisher Marine Research John Gauvin, Alaska Seafood Cooperative Tim Loher, Int. Pacific Halibut Commission Paige Drobny, Spearfish Research Andrew Seitz & Michael Courtney, U of Alaska Fairbanks, SAFS Suresh Sethi, Alaska Pacific Univ. and Cornell

Funded by grants from the North Pacific Research Board and NOAA’s Saltonstall-Kennedy program

Bering Sea flatfish (all Pleuronectidae)

Biomass 4.9 million mt (2015 AFSC shelf survey)

• Yellowfish sole Limanda

aspera

• Northern rock sole

Lepidopsetta polyxystra

• Flathead sole

Hippoglossoides elassodon

• Alaska Plaice

Pleuronectes quadrituberculatus

• Arrowtooth flounder

Atherethes stomias

• Pacific halibut

Hippoglossus stenolepis

Also: Greenland turbot, Kamchatka flounder, rex sole, Dover sole, Bering flounder, Sakhalin sole, butter sole, longnose dab, starry flounder

• 225,000 - 250,000 mt / year recently
• Catcher / processor vessels
• Quotas allocated by vessel
• Fixed halibut bycatch mortality

limits - allocated between vessels

• Significant constraint
• Recently reduced
• Catches sampled entering

processing area (after holding)

Other Flatfish Flathead Sole Arrowtooth Flounder Alaska Plaice Rock Sole Yellowfin Sole

Trawl fisheries for flatfish in the Bering Sea

Pacific halibut (bycatch, must be released)

(NMFS AKR)

Trawler deck-sorting to reduce halibut mortality

• Deck-sort to release bycaught

halibut as soon as possible

• Much lower mortality than if

put into holding tanks

• Current survival estimates from
• nboard viability assessments
• Structured set of reflex and injury
• bservations
• Scored Excellent, Poor, ‘Dead’
• Viability survival rates

(E-80%, P-45%, D-10%) based on 1970 tag releases

• Vessel specific mortality

estimates for trials

Project Goals

• Apply modern tag technology to validate and

improve the foundation for halibut survival rates

• Develop tag tools and metrics to monitor

halibut survival (Julie Nielsen’s earlier talk)

• Compare viability-based survival outcomes with

those indicated by recorded halibut activity

• Find conditions that improve survival
• Guide further survival improvements

Deployment Trips

Trawl - June Trawl - June Trawl - Aug Longline – Jul1

• Pac. Ocn. Perch

POP_2 (8) Flounder, turbot ATF/GTR_2 (25) Flathead sole FHS_1 (34) Yellowfin sole YFS_1 (39) Halibut longline (10) YFS_3 (45)

Tags on: 160 trawl releases 20 Controls

Halibut handling and data

• Stratified halibut selection – Time on deck
• 1 per 5 minute intervals – random time within
• Video viability assessments
• Reflex and injury scan, scored Excellent, Poor, or ‘Dead’
• Scored from videos by 5 experienced observers
• Fish and handling data recorded
• fish length
• time-on-deck
• tow duration
• deck temperature
• seafloor temperature
• humidity
• sand in catch
• sea state
• trawl net
• excluder use

Camera

Tags popped up and reported

• Data reported from

all 180 tags! (only

• ne with too little

data to use)

• Some appeared to

release from active fish (anchor failure?)

• Most recoveries in

vicinity of release, at all durations

• Some longer

movements from full

• r near-full

deployment periods

0 – 4.9 days 5 – 14.9 days 60+ days 15 – 59.9 days

Classification of activity data:

live or dead and when

Alive all 60 days Died after 3 days

• Scored by 6 reviewers
• Majority outcome after reconciliation discussion

(Nine dropped – still tied or too uncertain)

Analysis methods

• Kaplan-Meyer survival curves – estimate probability
• f surviving by time since release (can account for

early releases)

Trip

• Cox proportional hazards modelling to regress

multiple factors against survival period.

Survival analysis

Nearly all deaths

• ccurred

in the first 2 weeks

Key features of survival curves

Viability scores D P E

Variability by Trip-Target

.01 .45 .80

Survival analysis

Compare survival estimates between tagging results and from viability scores

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

ATF/GTR_2 FHS_1 POP_2 YFS_1 YFS_3

Viability Survival Tag Survival

Most Trip-Targets have reasonable matches (POP only 8 fish)

Trip 3 (YFS) high viability, Low tag survival

Survival by trip and viability

Excellent Poor ‘Dead’

• Most trip/targets – Survival rates by

viability score similar to each other and standard rates

• YFS Trip 3 – Survival rates for

excellent and poor scores substantially lower than other trips and standard rates

Surface/air temperatures

Third trip hypotheses

Increased mortality due to: Assessment handling Deck handling

Third trip hypotheses

Viability assessments overestimated: Pre – assessment recovery tank

Multiple potential differences unique to Trip 3 precluded isolating a single cause for the low survival

• f released halibut in spite of scoring high viabilities

Factors predicting mortality

Handling, fish size, environmental Cox proportional hazards model

Measurement pvalue AIC import. SpecimenLength 0.003 97% OnDeckTime 0.007 92% Trip1-FHS 0.862 96% Trip2-POP 0.149 Trip1-YFS 0.004 Trip3-YFS 0.012 TowDuration 0.010 92% Sediment in catch n.s. 27% DeckTemp* n.s. 37% CatchSize n.s. 26%

* - Also Surface Temp, Surface Temp – Bottom Temp (all highly correlated)

Factors predicting mortality

Model comparison (132 halibut with viabilities)

Model structure

Npars AIC R2

Without Viability

DaysSurvive ~ All factors except Viability 10 640 0.47 DaysSurvive ~ SpecimenLength 1 652 0.20 DaysSurvive ~ OnDeckTime 1 653 0.19 DaysSurvive ~ Trip/Catch 4 662 0.16 DaysSurvive ~ TowDuration 1 662 0.09 DaysSurvive ~ SpecimenLength + OnDeckTime + TowDuration 3 641 0.34 DaysSurvive ~ SpecimenLength + OnDeckTime + TowDuration + Trip/Catch 7 637 0.45

With Viability

DaysSurvive ~ All factors including Viability 12 591 0.74 DaysSurvive ~ Viability 2 623 0.47 DaysSurvive ~ Viability + SpecimenLength + OnDeckTime + Tow Duration 5 601 0.64 DaysSurvive ~ Viability + SpecimenLength + OnDeckTime + TowDuration + Trip/Catch 9 591 0.72

Next steps

• Better understand variation across

fisheries/seasons/vessels

• Viability scoring from deck sorts
• More tagging? How to distribute?
• Explore survival monitoring options
• Viability? Length-TOD-Tow time? Both?
• How much variability is acceptable?
• Does monitoring motivate practices that increase

survival?

• Further tag developments for survival use

Summary

• Acceleration-only tags measure survival and

tell us a lot about what happens to halibut after deck release

• Deck sort process greatly improves halibut

survival (Time on deck <25 min. vs. hours)

• Viability predicts survival well – except for

trip 3 – Could not isolate specific cause

• Fish size, time-on-deck and tow time all

influence survival

• Trip and Fishery variability is also significant

Captains and crews of: F/V Constellation F/V Cape Horn F/V Arica F/V Darlynne

Viability reviewers Elanor Wolfe Nicholas Rubino Laura Mowczan Alisha Foster Christopher Kromm

• Thanks to:
• Coauthors!
• Sarah Webster

Questions? Discussion

5 10 15 20 25 50 60 70 80 90 Number Length (nearest 10 cm) ATF/GTR_2 (25) FHS_1 (22) POP_2 (8) YFS_1 (36) YFS_3 (41)

Length composition of tagged halibut

15 December 2017 IFS 10 21