Horse mackerel 9a long-term Management Plan MSE analysis M Azevedo - - PowerPoint PPT Presentation

Horse mackerel 9a long-term Management Plan MSE analysis

M Azevedo(1), H Mendes(1), G Costas(2), E Jardim(3), I Mosqueira(3), F Scott(3)

(1) IPMA-Lisboa; (2) IEO-Vigo; (3) JRC-ISPRA

PELAC meeting Denn Haag, 11 July 2017

Outline

• Process (roadmap)
• Stock assessment

data/method: 1992-2016; Statistical catch-at-age with AMISH (settings: hom9a stock annex)

• Management Strategy Evaluation (MSE)

HCR and management options following questionnaire and meetings with stakeholders

2

3

• Results

Short-term (ST): 2017-2020; 2017-2027

Long-term (LT): 2070-2080; 2017-2080 for Precautionary considerations

Catch

Median; P10%, P90%, Interannual Variation (IAV) IAV1: mean (catchy/catchy-1) IAV2: sum (catch y-catchy-1) IAV3: mean (catch y-catchy-1)

Fishing mortality (F)

Median F P(F > FMSY)

Spawning Stock Biomass (SSB)

Median P(SSB < Blim) P(SSB < MSY Btrigger)

• Robustness/Sensitivity

Selectivity, Status quo period, Productivity

Focus group meetings on Management Plan for hom9a

4

• Oct 2014, SWWRAC/PELAC meeting, Madrid

First debate on Management plan

• Nov 2014, SWWRAC/PELAC webex meeting

Type of harvest control rule (TAC, F or Harvest Rate)

• Feb 2015, SWWRAC meeting, Lisbon

Refinement of HCR type and Biological Reference Points

• Jul 2015, PELAC meeting, Denn Haag

Preliminary estimates for horse mackerel (hom9a) Biological Reference Points - BRPs (WD)

Process

5

• Feb 2016, PELAC meeting, Denn Haag

BRP estimates (2015 assessment data); Rationale accepted by PELAC

• Mar 2016, SWWRAC stakeholders meeting, Matosinhos

Stakeholders feedback on options for catch stability; Level of catch for Fby-catch

• Oct 2016, PELAC meeting, Denn Haag

BRP estimates (2016 assessment data) accepted by ICES for advice; results from 1st set of stochastic simulations (MSE short-cut approach); HCR and BRPs acepted by PELAC; questions to stakeholders on assumptions & MP options -> produce questionnaire to be sent to stakeholders

• Nov 2016, SWWRAC/PELAC meeting, Lisbon

Synthesis of stakeholders response to questionnaire; set rodmap for further analysis

• Jul 2017, SWWRAC/PELAC meeting, Matosinhos

Preliminary results from stochastic simulations using full MSE; stakeholders feedback on HCR, options and metrics

• Jul 2017, WGHANSA meeting, Bilbao

Feedback from scientists on full MSE

Feb 2017: Benchmark hom9a WKPELA Jul 2016: ICES Stock Assessment WGHANSA

MSE

• Operating Model
• Productivity (Recruitment)
• Biological characteristics (natural mortality, mean weight,

proportion mature)

• Starting conditions: assessment with data 1992-2016
• Management Procedure

– Statistical catch-at-age (SCA: a4a) stock assessment (mimic AMISH) with observation error – Advice from short-term forecast (as in hom9a stock annex) – Management = Advice, fully implemented usually designated by full MSE

7

Management scenarios

(0.11) (0.01) (103 th t) (181 th t)

 FMSY target year: 2025  TAC: TACy+1=Catchy-1 TACy+1=mean[Catchy-3:Catchy-1] TACy+1=TACy

 TAC change limit: No; ± 15% and ± 20% catch constraint

8

Fishing mortality Spawning Biomass

FMSY Fby-catch Blim Btrigger HCR

Simulations

• 200 populations projected until 2080
• 2 productivity scenarios:

geometric mean hockey-stick (BRPs)

9

200 400 600 800 2000 4000 6000 8000 12000

SSB-R with 90%CI

SSB recruits

Results

10

• AMISH (current assessment method) & SCA (a4a)

Correlations: in the range 0.71- 0.95; high in most recent years which are initial conditions for simulations

11

Hockey-stick; FMSY in 2025; option catch ± 15%

TACy+1=Catchy-1

TACy+1=mean[Catch3 yrs] TACy+1=TACy No change limit

y: assessment interim year

blue shaded area: 80% confidence interval; blue line: median values green line: illustrates the trajectory of one of the 200 simulated populations million th t th t

12

Zoom on catch stability over time: option catch ± 15%

2017-2020: IAV1 = 10% IAV 2 = 58 t IAV3 = 15 t 2017-2020: IAV1 = 2% IAV 2 = 32 t IAV3 = 8 t 2017-2020: IAV1 = 4% IAV 2 = 25 t IAV3 = 6 t 2017-2020: IAV1 = 7% IAV 2 = 75 t IAV3 = 19 t

13

True F and Perceived F: option catch ± 15%

FMSY Perceived F is higher than realized (true) F  realized F doesn´t reach FMSY

14

Hockey-stick; FMSY in 2025; option catch ± 20 %

TACy+1=Catchy-1

TACy+1=mean[Catch3 yrs] TACy+1=TACy No change limit

15

True F and Perceived F: option catch ± 20% Very similar to

• ption catch ± 15%

IAV1: around 1% max IAV2: 5t max IAV3: 2t

16

METRICS TACy+1=Catchy-1 TACy+1=mean[Catch3 yrs] TACy+1=TACy No change limit ST 2017-27 LT 2070-80 ST 2017-27 LT 2070-80 ST 2017-27 LT 2070-80 ST 2017-27 LT 2070-80 Catch Median catch

47 41 45 41 51 41 48 40

10th perc. Catch

37 34 38 34 41 34. 37 32

90th perc. catch

57 50 53 50 61 50 68 49

IAV1

5% <1% 5% <1% 2% <1% 6% <1%

Fishing mortality P (F > FMSY)

2% 3% <1% 4% 23% 2% 29% 1%

Median F

0.105 0.107 0.09 0.107 0.114 0.104 0.113 0.101

Precautionary considerations Median SSB

459 347 486 345 425 352 412 367

P (SSB < Blim)

0% 0% 0% 0% 0% 0% 0% 0%

P (SSB < Btrigger)

0% 0% 0% 0% 0% 0% 0% 0%

Hockey-stick; FMSY in 2025; option catch ± 15%

Catch and SSB in thousand tonnes

2017-2080

17

METRICS TACy+1=Catchy-1 TACy+1=mean[Catch3 yrs] TACy+1=TACy No change limit ST 2017-27 LT 2070-80 ST 2017-27 LT 2070-80 ST 2017-27 LT 2070-80 ST 2017-27 LT 2070-80 Catch Median catch +2 +2 10th perc. Catch

• 3
• 3
• 2

90th perc. catch +5 +2 IAV1 Fishing mortality P (F > FMSY) +1 +2 Median F <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Precautionary considerations Median SSB

• 1

+0.01 +0.01

• 0.01

<0.01 +0.01 P (SSB < Blim) P (SSB < Btrigger)

± 20% vs ± 15%

values in % 2017-2080

18

• Selectivity accomodates pop structure over time
• P(SSB < Blim) = 0
• Max median Catch ≤ 12%
• Max IAV1 = 1%
• Status quo (3 year mean)
• P(SSB < Blim) = 0
• Max median Catch ≤ 5%
• Max IAV1 <1%

Robustness

19

Recruitment (low=40% geometric mean)

FMSY in 2025; option catch ± 15%

2017-2080: P1=42/80% P2=81%

P1=P(SSB < Blim) P2=P(SSB < MSY Btrigger)

2017-2080: P1=56/78% P2=80% 2017-2080: P1=0/0% P2=40% 2017-2080: P1=0/0% P2=20% MSY Btrigger Blim

Average probability across 200 populations Maximum probability over the 200 populations

20

Recruitment (low=40% geometric mean)

FMSY in 2025; option catch ± 20%

2017-2080: P1=5/30% P2=71%

P1=P(SSB < Blim) P2=P(SSB < MSY Btrigger)

2017-2080: P1=12/52% P2=73% 2017-2080: P1=0/0% P2=37% 2017-2080: P1=0/0% P2=20% MSY Btrigger Blim

P1 decreases in relation to ± 15%

21

Summary of results

• The harvest control rule proposed by stakeholders for the Long-Term Management

Plan for Southern Horse Mackerel (hom9a), with FMSY (0.11) target year in 2025, Blim=103 th t and MSY Btrigger=Bpa=181 th t, is precautionary for all the management scenarios explored.

• There are marginal differences between the options 15% and 20% TAC change

limits.

• The LT catch is estimated to be around 40 000 t (median value) but in the short-

term the IAV in catch is smaller for the management option TACy+1=TACy , of 2% (between 5-6% for the other options).

• Sensitivity tests indicate that the HCR combined with catch stabilizing mechanisms

is robust to changes in selectivity and status quo peridod. In a scenario of low recruitment, only catch options “TACy+1=TACy“ and “no TAC change limit”, are precautionary. With 20% change limit there is lower prob of falling below Blim in all scenarios.

22

Follow up ?