Mass Marking and Mark-Selective Fisheries

1
Mass Marking and Mark-Selective Fisheries

• Black and White Choice, or Complex Shades of Grey?

David Hankin Department of Fisheries
Biology Humboldt State University
2
Background Perspective (DGH)

• Hatchery Reform Efforts are to be Commended and
  are Especially Needed at CV Hatcheries (Hankin
  1982, Asilomar conference).
• Mass Marking (MM 100 AD-clip some smaller
  fraction with CWT) is not clearly a hatchery
  reform measure.
• Its hard to imagine implementation of MM (as
  defined above) without Mark-Selective Fisheries
  (MSF).
• MM MSF raise very serious issues.
• Other CV hatchery reform efforts seem much more
  critical (e.g., eliminate off-site releases
  Klamath vs CV fall Chinook genetic differences
  among stocks reconsider random mating)

3
Coordination of Presentations

• Hankin (HSU) Overview of hatchery marking
  programs Intention of MM serious issues raised
  by MM MSF (PSC Expert Panel) DIT? Estimation
  vs Imputation
• OFarrell (NMFS/PFMC) - MMMSF Assessment and
  Management Issues ESA-listed stocks Disaster
  Prevention?
• Phillips (CDFG) Logistics, costs and
  feasibility of recovery programs under MM MSF
  

4
Historical Perspective on Hatchery Marking
Programs for Chinook Salmon

• Individual groups of fish were tagged (fin clips)
  primarily to assess relative performance of
  different rearing and release strategies.
• No serious attempts or apparent interest in
  marking to allow identification of hatchery fish
  or separation of wild from hatchery fish.
• High variability in marking programs across years
  within hatcheries and across hatcheries
  generally impossible to reliably estimate
  hatchery contribution to returns.

5
Development of Coded-Wire Tags

• In mid-70s, AD-CWT combination, combined with
  rigorous ocean and freshwater escapement sampling
  programs (e.g., Klamath R.) allowed cohort
  reconstruction/analysis.
• Cohort Analysis Fishery exploitation rate
  analysis survival rates maturation rates but
  not hatchery contribution due to variability in
  marking levels.

6
By 2004 CWT System Summary statistics (Johnson,
K. 2004)

• 54 state, federal, tribal and private entities in
  USA and Canada conduct CWT experiments 1200 new
  codes annually
• gt 50 million juvenile salmon and steelhead tagged
  annually (cost gt 7.5 million)
• About 275,000 CWTs recovered each year (cost
  12-13 million)
• Used for many purposes evaluate success of
  hatchery practices ocean distributions serve as
  proxy for fishery impacts on unmarked wild fish

7
Key Assumptions and Requirements of Coastwide CWT
Recovery Program

• Adipose fin clip sequestered as a mark to be used
  ONLY for presence of CWT
• Indicator Stock Idea Fate (fishery impacts) of
  hatchery fish with ADCWT assumed to be same as
  that of nearby wild population with similar
  life history, run timing
• ADCWT releases were not designed to allow
  estimation of overall contribution of hatchery
  fish to escapement ? Hatchery fish unknown.

8
Key Advantage of CWT Releases

• Enrichment of small populations in ocean catch
  sampling
• Assuming equal survival of two 200,000 fish
  release groups, expected ocean catches of CWTd
  fish should be similar even when size of
  populations from which CWT groups originated may
  differ by order of magnitude
• Very advantageous for assessment of small natural
  populations (assuming appropriate proxy).
  Example Sacto winter Chinook.
• NOTE MM (100 AD-clip) eliminates enrichment.

9
CV Fall Chinook Hatchery ADCWT Marking Programs
circa 2000
Hatchery Prod. Goals (Mil) Size at Release Release Locations Tagged BY
Merced 0.96 smolts/ or 0.30 year. 70-90/lb 6-10/lb Merced San Joaquin 63 1996 97 1994 43
Moke. 3.25 smolts 1.5 year. 2.0 ocean 40-75/lb 5-8/lb 25-30/lb Moke. R. SF Bay 17 11 1995 13
Feather 6.0 smolts 2.0 ocean 40-60/lb 25-30/lb Pri. SF Bay SF Bay 32 1996 11
Nimbus 4.0 smolts 40-60/lb SF Bay 1996 0
Coleman 12.0 smolts 60-80 mm Pri. Battle Ck. 08 08
10
Reliable Estimation of the Proportion of Hatchery
Fish in Spawning Runs Constant Fractional
Marking (Hankin 1982)

• Allow existing ADCWT programs to continue.
• Mark a constant fraction of remaining
  production releases with fin clip.
• c1/(fraction marked)

11
Implementations of CFM

• 1979-1982 BYs Iron Gate and Trinity River
  hatcheries (Klamath system) fin-clip 1/3 of all
  releases in excess of ADCWT (Hankin 1982)
• 1999 BY present Trinity River Hatchery 25
  of all releases receive ADCWT (Hankin Newman
  1999)
• 2007 BY present CV hatcheries 25 of all
  releases in excess of experimental CWT groups
  receive ADCWT (Hicks, Newman and Hankin 2005)

12
Trinity R. Chinook Willow Creek Weir 2008
Julian Grilse Grilse Adults Adults
week Inclusive dates Inclusive dates Inclusive dates Total Ads Total Ads
34 20-Aug 20-Aug 26-Aug 120 1 81 4
35 27-Aug 27-Aug 2-Sep 132 1 64 2
36 3-Sep 3-Sep 9-Sep 203 5 87 3
37 10-Sep 10-Sep 16-Sep 282 9 170 14
38 17-Sep 17-Sep 23-Sep 126 5 122 12
39 24-Sep 24-Sep 30-Sep 143 8 146 25
40 1-Oct 1-Oct 7-Oct 39 4 77 12
41 8-Oct 8-Oct 14-Oct 17 1 47 6
42 15-Oct 15-Oct 21-Oct 27 2 84 14
43 22-Oct 22-Oct 28-Oct 18 0 39 4
44 29-Oct 29-Oct 4-Nov 9 0 89 15
2008 Totals 1,134 36 1,009 111
13
Purposes of 100 Marking

• Historical Provide fishing opportunities on
  hatchery fish (e.g., steelhead) while reducing
  impacts on wild fish N. Umpqua, OR steelhead -
  If hes wild, let him go.

14
Other Purposes of 100 Marking

• Contemporary theme
• To allow control of hatchery fish among
  hatchery spawners and among naturally spawning
  fish in an integrated hatchery/natural population
  system (HSRG 2004)
• Note Implicitly assumes on-site releases of
  hatchery fish (to reduce straying across
  populations) and system of weirs or similar
  structures to allow separation of types on
  natural spawning grounds. Is that likely in the
  CV in the near future???

15
Mass Marking and Mark-Selective Fisheries
Implementation for Pacific Salmon

• In current mass marking procedures, ALL hatchery
  fish receive an adipose fin clip, but many (most)
  of these fish are released without CWTs
• (Norm Dicks, D-Wash, 2003) The United States
  Fish and Wildlife Service shall, in carrying out
  its responsibilities to protect threatened and
  endangered species of salmon, implement a system
  of mass marking of salmonid stocks, intended for
  harvest, that are released from Federally
  operated or Federally financed hatcheries
  including but not limited to fish releases of
  coho, chinook, and steelhead species. Marked fish
  must have a visible mark that can be readily
  identified by commercial and recreational
  fishers."

16
Achieving mass marking has required development
of new (and expensive) automated tagging
technologies
AutoFish system from Northwest Marine Technology
is capable of automated removal of the adipose
fin and/or insertion of a CWT. Each fish is
sorted by size, clipped and/or tagged, and
returned to the pond in about 5 seconds.
17
Mass Marking also introduces new complications
wrt sampling catches and escapements for CWTd
fish because now not all ad-clipped fish have CWT
and it would be foolish to collect heads from all
AD-clipped fish to search for non-existent CWTs.
Tube (left) and Wand (right) CWT detectors
18
More complications from MM and MSF

• In mark-selective fisheries (MSF), only
  ad-clipped (known hatchery) fish may be retained,
  theoretically leading to reduced fishing
  mortality on weak (e.g.
• ESA-listed) natural stocks.
• BUT there must also be some non-catch hook
  release mortality on unmarked (natural) fish in
  MSF.
• Recovery patterns of CWTd hatchery indicator
  stocks have routinely been used to infer
  exploitation rates experienced by associated
  natural stocks.
• BUT with MSF, recovery patterns of marked
  hatchery fish are no longer the same as for the
  associated unmarked natural stocks.

19
PSC Expert Panel Findings (2005)

• 6. Mass Marking (MM) and Mark-Selective Fisheries
  (MSF) together pose serious threats to the
  integrity of the CWT system. The PSC has been
  alerted to these threats since at least 1991. In
  particular
• A. Recovery patterns of adipose-clipped fish no
  longer indicate recovery patterns for unmarked
  natural stocks (see OFarrell talk)
• B. Significant practical and statistical issues
  are raised by the need to find AdCWT fish when
  many MM fish are released with Ad clip only. (see
  Phillips talk)

20
These Concerns are not new

• 1991 letter to Director General, Pacific Region,
  and Director (WFDW) noted 7 different significant
  issues (problems) that were raised by MM and MSF.
  (See EP Report)
• Letter was sent by Co-Chairs of PSCs CTC
  (Chinook Technical Committee
• Jim Scott (WDFW), Norma Sands (US), Brian
  Riddell (DFO)

21
DIT to the Rescue???Double Index Tagging Two
groups receive CWT, but only one group is
AD-clipped.
22
PSC Expert Panel Findings (2005)

• 7. For coho and Chinook salmon, it appears
  possible to estimate total non-catch mortalities
  at age in all MSFs from a full cohort analysis of
  paired DIT (double-index tagged) releases, IF
  complete recovery data can be collected for DIT
  groups.
• However, we could not derive an unbiased method
  to allocate total non-catch mortalities over a
  set of MSFs. It is therefore impossible to guess
  where and when non-catch impacts take place.

23
Hankins Main Concern

• In the absence of successfully executed DIT
  experiments (there have been none so far for
  Chinook of which I am aware see Expert Panel
  report), non-catch mortalities to natural stocks
  would have to be imputed based on a sequence of
  suppositions and assumptions. This is not
  equivalent to estimation from observed data!

24
Imputation Vs Estimation Whats the Difference?

• Imputed Non-Catch Mortalities
• - Assume (i.e., make up)
• Known contact rate
• Known shaker mortality rates (sport vs comm.)
• Independence of mortality if multiple releases of
  same fish
• Estimated N-C M (i.e., let data speak)
• Statistical analysis of recovery data (for DIT
  groups?)

25
An Extreme Example of Imputation

• Guessing the probability of a nuclear plant
  meltdown (as in guessing non-landed mortalities
  in MSF)
• Assume that all events leading to a meltdown are
  statistically independent of one another ?
  meltdown is highly improbable vs
• But, what if events are not independent, but one
  failure causes other failure events to be more
  likely ? not so improbable
• Frequentist approach (based on observed metdowns)
  is unsuitable for meltdowns (i.e., wait for
  meltdowns?), but is feasible for fish! Guessing
  is not acceptable if there is a better way.

26
MM Asking the Hard Questions

• Q. If the purpose of MM is not to support MSF,
  but to support hatchery reform, then why remove
  AD fins from all fish?
• A. Removal of AD fin is not needed to develop
  integrated hatchery programs. Instead, say, 25
  could be ADCWT for assessment purposes and
  remaining unmarked fish could receive blank wire.
  On return to FW, tubes (hatcheries) and wands
  (weirs) could be used to separate wild from
  hatchery fish (M. Mohr).

27
MMMSF Asking the Hard Questions

• Observation For a number of reasons, especially
  for commercial fisheries, MSF only makes sense,
  theoretically, when there is a high proportion of
  tagged fish among those likely to be contacted by
  a fishery (say, gt 70). (Commercial fishermen
  have formally complained about MSF in PSC ocean
  fisheries with low mark rates.)

28

• Follow-up Q.
• For CV Chinook, what kind of future ocean
  contribution do we want to see from naturally
  produced as compared to hatchery fish?
• Does it make sense to promote a fishery type that
  makes no sense in the context of the future that
  we would like to see?
• How easy would it be to stop a large-scale
  implementation of MSF in Central CA ocean
  fisheries once it had started?

29

• THE END