The assessment(s) Data used (and not used): Catch data Surveys - - PDF document

▶
the assessment s data used and not used
SMART_READER_LITE
LIVE PREVIEW

The assessment(s) Data used (and not used): Catch data Surveys - - PDF document

Nov 09, 2023 162 likes •247 views

The assessment(s) Data used (and not used): Catch data Surveys Biological reference points David Miller The 2010 assessment Wageningen IMARES Potential issues/uncertainties in the assessment Way forward? WGWIDE

slide-1
SLIDE 1

David Miller

Wageningen IMARES

The assessment(s) Data used (and not used):

  • Catch data
  • Surveys

Biological reference points The 2010 assessment Potential issues/uncertainties in the

assessment

Way forward? WGWIDE examines 3 models:

  • TISVPA: separable selection (>1994)
  • XSA: non3separable 3 assumes perfect catch data
  • SMS

SMS SMS SMS: Stochastic Multi3species Model (separable)

All analytical age3structured assessments

  • Ages 1310+ (age 10 is a plusgroup)

No benchmark assessment

  • Based on previous evaluations and comparisons,

SMS has been chosen as the final assessment for the last 5 years

Catch values are provided to ICES by the

member nations fishing the stock

Estimates are considered to be reliable

  • Constant selection pattern for the

catch 2 periods: 1981–1999,1999–2009 First age with age independent catchability 8 Age groups with the same variance 1, 2, 3–6, 7–10

International Blue Whiting Spawning Stock Survey

International Blue Whiting Spawning Stock Survey International Blue Whiting Spawning Stock Survey International Blue Whiting Spawning Stock Survey

Norway, Russia, the Faroe Islands and the EU Spawning grounds west of the British Isles March3April (peak of spawning)

  • 2004'2010

First age with age independent catchability 5 Age groups with the same variance 3–8, min std 0.4

  • 18
  • 16
  • 14
  • 12
  • 10
  • 8
  • 6
  • 4
  • 2
2 4 Longitude 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 Latitude Tridens Celtic Explorer G.O. Sars Vimus Magnus Heinason Cruise tracks 50.0° 51.0° 52.0° 53.0° 54.0° 55.0° 56.0° 57.0° 58.0° 59.0° 60.0° 61.0° 62.0° 63.0°
  • 20°
  • 19°
  • 18°
  • 17°
  • 16°
  • 15°
  • 14°
  • 13°
  • 12°
  • 11°
  • 10°
  • 9°
  • 8°
  • 7°
  • 6°
  • 5°
  • 4°
  • 3°
  • 2°
  • 1°
200m 500m 1000m Faroes Netherlands Ireland Norway Russia

Current survey plan Proposed plan for 2011

slide-2
SLIDE 2

Primary source of information about the current

state of the stock

Covers a large area

  • Not always well standardised
  • Limited time
  • Migrations occur during the survey
  • Weather conditions often unfavourable

Short time series

  • ‘catchability’ estimates still variable
  • In a transition phase, should stabilise in the future

International Ecosystem Survey in the Nordic

International Ecosystem Survey in the Nordic International Ecosystem Survey in the Nordic International Ecosystem Survey in the Nordic Seas Seas Seas Seas

EU, Norway, the Faroe Islands and Russia Barents Sea, north & central Norwegian Sea

and southwestern Norwegian Sea (Faroese and Icelandic Zones)

May

  • ,

2000'2010 First age with age independent catchability 2 Age groups with the same variance 1,2

Trawls Stations Blue Whiting is not the primry target of the

survey

Generally small and loose registrations of

blue whiting

Used as an index of recruitment (incoming

year3class strength)

Norway Spawning grounds west of the British Isles Precursor for IBWSSS 199132006

  • 200432006 incorporated in IBWSSS
  • 2007 onwards: Norwegian contribution to IBWSSS

changed

  • ,

1993–2003 First age with age independent catchability 5 Age groups with the same variance 3–4, 5–6, 7–8

Norway, some years in co3ordination with

Russia

Barents Sea late January‐early March Only used in forecast

Only used in forecast Only used in forecast Only used in forecast

  • Coverage on edge of the distribution area
  • Considered first reliable indication of year3class

strength (recruitment)

slide-3
SLIDE 3

NOT USED IN THE CURRENT ASSESSMENT

NOT USED IN THE CURRENT ASSESSMENT NOT USED IN THE CURRENT ASSESSMENT NOT USED IN THE CURRENT ASSESSMENT

Only representative for a small part of the

stock:

  • Spanish bottom trawl survey
  • Portuguese bottom trawl survey

Patchy temporal coverage / not updated:

  • Norwegian Sea summer survey (1981–2001, 2005–2007)
  • Faroes plateau spring bottom trawl survey (1996–2008)
  • Faroes plateau autumn bottom trawl survey (1994–2008)

Spanish pair trawls CPUE

  • 198332003
  • Discontinued because fleet only represents a small

part of the landings in a small part of the distribution area

Norwegian CPUE

  • 198232003
  • Not updated
  • Not considered representative of the development
  • f the stock
Type Name Year range Age range Variable from year to year Yes/No Caton Catch in tonnes 1981 – 2009 1'10 Yes Canum Catch at age in numbers 1981 ' 2009 1'10 Yes Weca Weight at age in the commercial catch 1981 – 2009 1'10 Yes West Weight at age of the spawning stock at spawning time. 1981 – 2009 1'10 Yes Mprop Proportion of natural mortality before spawning 1981 – 2009 1'10 No Fprop Proportion of fishing mortality before spawning 1981 ' 2009 1'10 No Matprop Proportion mature at age 1981 ' 2009 1'10 No Natmor Natural mortality 1981 ' 2009 1'10 No Tuning data: Type Name Year range Age range Tuning fleet 1 Norwegian Acoustic Survey 1991'2003 3'8 Tuning fleet 2 International Ecosystem Survey 2000 ' 2010 1'2 Tuning fleet 3 International Spawning Stock Survey 2004 ' 2010 3'8

History:

  • SGPA 1998:

Blim = 1.5 million t (~lowest observed at the time) Flim=Floss=0.32 Bpa (2.25 Mt) and Fpa (0.21) arbitrarily calculated up from these Simulations suggested in absence of clear stock3recruit relationship

  • ACFM 1998

Fpa=0.32 (avg. F, no apparent neg. effects on rec.) Flim raised to 0.51

  • SGPRP 2003:

Bloss=1.2 Mt, similar to current Blim, assessments still unstable – no change History:

  • ACFM 2006

Using Bpa as a trigger expected to lead to a >5% risk

  • f SSB dropping below Blim

Suggests Bpa should be higher

  • WKREF 2007

In the period of low recruitment there seems to be no sign of reduced recruitment at low SSB In the period of high recruitment no apparent trend in recruitment over the range of biomass May be no need for different Blim values in different productivity regimes

History:

  • MP simulation evaluation 2008

high risk of stock collapse for F > 0.3 F0.1=0.18, low risk to stock

Current BRPs:

Reference Point B B F F Value 1.5 mill t 2.25 mill t 0.51 yr'1 0.32 yr '1 Basis B Blim*exp(1.645* σ) With σ = 0.25 F F
slide-4
SLIDE 4

WGWIDE – assessment working group on

widely distributed and highly migratory stocks

Vigo, Spain Aug/Sep 2010 Participants include scientists from almost all

countries fishing the stock

Catch proportion at age for Blue whiting year age 2 4 6 8 10 1980 1985 1990 1995 2000 2005 2010 al

Area Season Deviations from pre3agreed survey plan:

  • Temporal: Russian component 2 weeks late

Excluded from survey (risk of ‘double counting’)

  • Spatial: Dutch component incomplete due to poor

weather

Gap in area coverage occurred in an area of concentrated fishing effort and thus likely to have contained a high but un un un un3 3 3 3quantified quantified quantified quantified biomass Interpolated (estimated) from surrounding surveyed areas WGNAPES considered the revised estimate to

be robust and recommended WGWIDE to use these values

  • 18
  • 16
  • 14
  • 12
  • 10
  • 8
  • 6
  • 4
  • 2
2 4 Longitude 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 Latitude Tridens Celtic Explorer G.O. Sars Vimus Magnus Heinason

Planned Realised

slide-5
SLIDE 5

2010 mean acoustic density the lowest observed

since 2004

  • 50% decrease in stock biomass compared to the 2009

survey

50° 54° 60° 62° 58° 52° 56° 0° 12° 20° 4° 4° 8° 16° sA - values 0 - 100 100 - 500 500 - 1000 1000 - 141000

2009 2010 Largest coordinated coverage in the

Norwegian Sea and surrounding waters in summer ever

Small and loose registrations of blue whiting Generally colder in the Norwegian Sea, but

extremely warm Atlantic water masses in the southern and southwestern part of Iceland

# International Norweigian Sea ecosystem survey 2000-2009 # Effort and catch numbers age 1 age 2 48927 3133 #2000 85772 25110 #2001 15251 46656 #2002 35688 21487 #2003 49254 22086 #2004 54660 19904 #2005 570 18300 #2006 21 552 #2007 29 75 #2008 0 14 #2009 0 0 #2010

Recruitment (age 1 yr) 10 20 30 40 50 60 70 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 Recruits, *10^9 SMS TISVPA XSA Fishing mortality 0.1 0.2 0.3 0.4 0.5 0.6 0.7 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 F 3-7 Spawning stock biomass 1 2 3 4 5 6 7 8 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 SSB (million t)

Close agreement in 3 models:

slide-6
SLIDE 6 1980 1985 1990 1995 2000 2005 2010 0.0 0.2 0.4 0.6 mean F 1980 1985 1990 1995 2000 2005 2010 2000 4000 6000 8000 SSB (1000 t) 1980 1985 1990 1995 2000 2005 2010 20000 40000 60000 Recruitment (10^6)

SMS

estimated Stock development with uncertainty bounds (95%)

  • Norw. Spawning Stock Surv.
1990 1995 2000 2005 2010 3 4 5 6 7 8 1.92
  • Intl. Surv. in Nord. Seas.
1990 1995 2000 2005 2010 1 2 3.29 IBWSSS 1990 1995 2000 2005 2010 3 4 5 6 7 8 0.69

Residuals Retrospective

fits

1980 1985 1990 1995 2000 2005 2010 2000 4000 6000 Retrospective anlysis: 2007 - 2009 SSB 1980 1985 1990 1995 2000 2005 2010 0.0 0.1 0.2 0.3 0.4 0.5 F 10000 30000 50000 recriuts 10^6

Over Over Over Overestimate? Under Under Under Underestimate?

IBWSSS

2010

  • Conclusion: The

2010 estimate is likely to be an underestimate underestimate underestimate underestimate

  • f the SSB
  • However this is

the only survey

  • nly survey
  • nly survey
  • nly survey

available covering the spawning stock

  • 2010
  • bservations are

important important important important for the assessment.

  • Bl. whiting

BW 2009 final run BW 2010, no IBWSSS 2010 BW 2010 all data

1985 1990 1995 2000 2005 2010 10 20 30 40 50 60 Recruits (billions) 1985 1990 1995 2000 2005 2010 0.0 0.1 0.2 0.3 0.4 0.5 F 1985 1990 1995 2000 2005 2010 2000 4000 6000 SSB (1000t)

IBWSSS 2010

  • Residuals

without 2010:

  • Residuals

with 2010:

IBWSSS 1990 1995 2000 2005 2010 3 4 5 6 7 8 0.45 IBWSSS 1990 1995 2000 2005 2010 3 4 5 6 7 8 0.69

IESNS

  • Bl. whiting
a) as input b) 0 to lowest observed c) survey excluded 2006 2007 2008 2009 2010 1 2 3 4 5 6 Recruits (billions) 2006 2007 2008 2009 2010 0.0 0.1 0.2 0.3 0.4 F 2006 2007 2008 2009 2010 1000 3000 5000 SSB (1000t)
slide-7
SLIDE 7

2010 assessment should be considered:

  • uncertain

uncertain uncertain uncertain with respect to the absolute estimates absolute estimates absolute estimates absolute estimates of stock metrics

  • certain

certain certain certain in the conclusion on the steep decline steep decline steep decline steep decline in both SSB and recruitment in the most recent years

The devil is in the data

data data data...

  • Assessment results highly dependent on the quality of the
  • nly survey that covers the spawning stock (IBWSSS)
  • Errors in the surveys may give misleading estimates of

stock abundance

  • Could attempt model formulations that better handle errors

in the data, or identify and ignore data that lead to wrong results

IBWSS is still short, at times inconsistent,

leading to high annual variability in assessment models fit to the data

  • Hopefully estimates of catchability will improve as

the series lengthens, reducing fluctuations of estimates in the terminal year

Two main areas of uncertainty:

  • Timing

Timing Timing Timing: aimed for peak of spawning, but this may vary over years (e.g. due to temperature)

  • Coverage

Coverage Coverage Coverage: earlier migration of the stock northwards can affect distribution at time of survey

  • More robust design for 2011 to try ensure complete

coverage

It is difficult to estimate the exact level of

recruitment in recent years, but there is no doubt that recruitment has been very low since 2006

IESNS recruit survey may no longer be

indicative of the strength of incoming year classes

  • In the most recent years there is a mismatch

between survey indices and the amounts caught from these year classes later on

  • Potentially small year classes may grow up

elsewhere

Benchmark assessment requested for the start of

2012

  • Not 2011 because time is needed to prepare work and

source potential new inputs or methods to present valid alternatives

  • Good platform to bring about assessment reforms

Industry information:

  • WGWIDE 2010 report: “No comprehensive information

has been received from the fishing industry this year”

How can industry contribute data or information to make the

assessment more robust?

Industry data to help improve our knowledge of the location of

juvenile fish?

Inform on center of abundance/timing of northward migration?