Lecture 11 review: Stock synthesis models

1
Lecture 11 review Stock synthesis models

• Synthesis model is a term coined by Methot for
  what had been called statistical catch at age
  (SCA) models examples are SS2, CASAL
• Basic idea is to use age-structured model to
  generate predictions of multiple types of
  observations
• Catch for multiple fleets with different age
  selectivities
• Length and age composition of catch
• Multiple abundance trend indices
• Basic aim is to reconstruct historical changes in
  stock size and recruitment
• Main limitation is bad trend index data and
  complex temporal change in size-age selection
  patterns

2
Parameter estimation and state reconstruction for
dynamic models
Observation errors
Process errors
Parameters
Observation Model (predicted y)
Data (observed y)
State dynamics Model N
Statistical criterion
y
N
Nt1Nt-Ct
ytqNt
Parameter q
Log-likelihood function
Parameter No
3
Lecture 12 topics dangerous quick fixes in
fisheries management

• Several simplistic solutions to management
  problems are defended with religious fervor by
  their proponents
• Stock enhancement
• Marine protected areas
• Individual Vessel Quotas
• Simplistic solutions derive from simplistic
  models (eg. produce more?catch more)
• Must take a systems view to understand why these
  solutions fail

4
Taking a systems view
Fishers and Other Stakeholders
Fish stocks, Ecosystem
Management (Assessment,Regulation)
These three subsystems are dynamically linked
mess with any one of them, and there will be
(sometimes pathological) responses in the other
two. Ignore such responses at your peril.
5
Hatchery programs the biggest single threat to
sustainable fisheries?

• Two types of hatcheries
• Production (meet growing demand by producing more
  fish)
• Conservation (breeding programs for the culls, so
  stupid as not to merit discussion)
• Based on concept that protected rearing can
  radically increase egg-juvenile survival rates,
  often by several orders of magnitude (e.g. from
  3 to 80 for pink and chum salmon)
• Huge growth in hatchery capability (species,
  efficiency) and capacity in last two decades,
  particularly in the North Pacific.

6
Huge growth in hatchery capacity North Pacific
salmon
From NPAFC 2007 Doc. 1060 (Malbec model report )
7
Hatchery programs the biggest single threat to
sustainable fisheries?

• Negative impacts on wild stocks
• Competition with wild juveniles leading to
  depressed juvenile survival
• Transmission of diseases, attraction of predators
• Fishing effort responses where gear takes both
  types of fish
• Genetic impacts low fitness of hatchery x wild
  crosses means hatchery fish in the wild can act
  like sterile male releases
• But there are situations where hatcheries have
  produced very large net benefits
• Systems with no wild spawning (e.g. BC lakes)
• Put and Take fisheries near population centers

8
When wild juvenile production already fills
available habitat, total catch may not increase
at all
9
Remedies for the dark side of enhancement programs

• Problem
• Poor survival after release
• Competition with wild juveniles
• Disease and predator impacts
• Attract fishing pressure
• Mate with wild fish (poor reproductive
  performance in wild)
• Declining hatchery performance over time

• Remedy
• Increase release size, acclimate at release sites
• Increase release size to avoid nursery area
  overlaps
• Prophylaxis, predator control programs
• Selective fisheries (marking, location)
• Two extreme options use only wild spawners, or
  pure captive brood stock
• No known remedy, mechanisms not understood

10
How scientists contribute to misunderstandings
about enhancement

• Myopic performance measures that are convenient
  to study (e.g. survival rates)
• Failure to assess competition and exploitation
  rate effects on wild stocks
• Focus on sexy topics (genetics) rather than the
  more important but difficult to study ones
  (disease, wild stock effects, effort responses)
• Value laden analysis and communication natural
  fish are intrinsically better, need for wild
  fish for risk management (diverse production
  portfolio)

11
Marine Protected Areas substituting mindless
protectionism for effective management

• Main arguments for MPAs
• Increased fishery yield in cases where of
  management fails to constrain efforts, especially
  on less valued stocks/species, through
  spillover effects
• Protecting 20 of area will insure spawning
  stocks at least 20 of unfished levels (Bohnsack)
• Protection of habitat from damage by fishing
  activities (especially trawling)
• Places where fauna can live undisturbed
  (basically an animal rights argument)
• Reference areas for assessing unfished
  abundances, longevity, etc.

12
Marine Protected areas the reality

• Fish move, so protecting 20 of the habitat
  typically protects much less than 20 of the
  stock full protection achieved only near center
  of very large MPAs
• Severe overfishing outside MPAs likely results in
  inadeqate larval seeding everywhere, including
  inside the MPAs (do not achieve natural abundance
  in them unless they are largely self-seeding, in
  which case they do not provide large spillover
  benefits anyway
• Wishful thinking networks of reserves will lead
  to connectivity where the reserves seed each
  other with larvae

13
Marine Protected areas the reality

• Empirical studies show abundances 2-4 x higher in
  reserves than outside (Halpern), but proponents
  of MPAs do not like to mention that most MPAs are
  sited in areas of higher abundance in the first
  place (this is dishonest science at its worst)
• Collateral damage to non-target stocks and
  habitat often very restricted in space/time
  exceptions like big trawled areas are treated as
  typical

14
Marine protected areas complex ecosystem
responses (McClanahan 2007)
Studies like this are revealing fairly strong
top-down effects of large predator recovery on
smaller species it is not obvious that reserves
even increase biodiversity in the long run
15
Marine protected areas the SLOSS debate

• Population and ecosystem models clearly indicate
  that SL (Single Large) is necessary to obtain any
  protection at all for more mobile species
• But it is clearly much easier from a social,
  economic, and political perspective to implement
  SS (Several Small)
• So the latest science advocacy game is to pretend
  that SS works, based on experience with coral
  reef MPAs

16
Marine protected areas where and when are they
actually needed?

• When there is no way to control efforts, and
  effort will remain high even at very low stock
  sizes (high prices, low cost, availability of
  productive species)
• When there is source-sink metapopulation
  structure should protect source or nursery areas
  that provide recruits to many other areas that do
  not self-seed
• When there is unacceptable damage to habitat or
  nontarget species with high existence value

17
New modeling approaches are offering guidance
about optimal mosaics of protected areas

• EDOM predicts long term responses of multiple
  populations, estimates optimum spatial
  distribution of fishing effort
• MARXAN uses GIS information to identify areas of
  high value by stakeholders with conflicting
  interests (fishing, protection) and to seek
  optimal spatial patterns
• ECOSPACE evaluates ecosystem-scale effects of
  alternative MPA proposals

18
Example optimized effort distributions from EDOM
Optimum effort when arbitrary closed areas (Plan
A) are imposed includes high efforts just at MPA
boundaries, but much more even effort
distribution absent such arbitrary closures the
optimized effort distribution for Plan A achieves
about 93 of the economic value that could be
achieved without MPAs.
19
IVQs and TURFs are the first step in getting the
incentives right for fishermen to cooperate with
scientists and managers to find sustainable
management solutions

• But just because we must have them doesnt mean
  that they will work right
• The main pitfalls are
• Setting the wrong Quota leads to depensatory
  fishing mortality rate
• Changes in fisher behavior when fishing with IVQ
  make historical cpue trend data unusable (must
  have surveys, direct U assessments, which cost a
  lot)
• Concentration of ownership the little guys
  always lose

20
Error propagation in stock size estimates makes
quota management very dangerous
There is a fairly high correlation between stock
size estimation error in year t1 and the error
in year t, i.e. errors persist (if overestimate
this year, will do so next year) for several
years (from Walters 2004 CJFAS 611061-1065)
21
How do you identify system-scale (indirect
effect) problems that may make some quick-fix
dangerous

• Keep your eyes open (think more broadly)
• Keep your ears open (listen to warnings from
  people who have been thinking about things that
  might go wrong it is the raving loonie you want
  to listen to most closely, not your trusted
  colleagues)
• Look closely at the things you might want to
  treat as constant parameters in models
  (survival rates, fishing effort, etc)
• Identify things you can control, and also things
  you cant