Why harvest

1
Lecture Outline Population Harvesting Ben
ONeal April 3rd, 2007

• Why harvest?
• Additive vs Compensatory mortality
• Maximum sustainable yield (MSY)
• Sustainable harvest strategies
• Age / Sex-structured harvest
• Harvest as a selective force
• Harvest as a management tool
• Adaptive harvest management (AHM)

2
Why harvest?

• Meat
• Income
• Recreation
• Management
• Land Conservation

3
Waterfowl

• 1.5 million hunters in U.S.
• 13 million waterfowl harvested/yr
• 1.6 billion/yr spent locally
• 25 million/yr for habitat from duck stamps

4
Additive vs. Compensatory Mortality
Does harvesting increase the overall mortality
rate for a population...
or simply remove a surplus of individuals that
would otherwise die?
5
Compensatory mortality requires density-dependent
survival
1
Additive
Natural survival rate
0.5
Compensatory
0
0
100
50
Population size
All of the issues regarding DD affect whether
hunting mortality is compensatory.
6
Example

• Assume that harvest mortality takes place first
  and then natural mortality occurs rest of year in
  density-dependent fashion.

Sn B0 B1N
where Sn is natural survival rate outside of
hunting season.
(Borrowed from Gary Whites Lecture Notes, CSU)
7

• Now, we remove hunting so that 90 individuals
  undergo natural mortality.

Sn 0.8333 0.005556 (90) 0.333
8
Types of compensatory mortality
1. Complete hunting mortality is completely
compensated for by increase in survival outside
of hunting season
2. Partial hunting mortality is partially
compensated for by increase in survival outside
of hunting season
9
Types of compensatory mortality
3. Threshold hunting mortality is compensated
for by an increase in survival outside of
hunting season to a threshold harvest value (c).
Beyond threshold, population cannot compensate
for harvest and overall mortality increases.
Additive
Compensatory
Annual survival rate (S)
Annual survival rate (S)
c
Hunting mortality rate (K)
Hunting mortality rate (K)
10
Empirical evaluations compensatory vs. additive
mortality

• Band recovery data from gt410,000 adult mallards

• Multiple studies in N. America from 1950 to 1979

• Alternative hypotheses complete compensatory
  and totally additive

• Rejected the hypothesis of total additivity and
  concluded that it appears that hunting
  mortalities are largely compensated for by other
  forms of mortality.

(Burnham, KP and DR Anderson. 1984. Ecology
65105-112)
11
Empirical evaluations compensatory vs. additive
mortality

• Band recovery data for mallards from 1979-1989.

• Strongly rejected the complete compensatory
  hypothesis.

• Concluded that under certain conditions,
  restrictive regulations can successfully increase
  survival rate of mallards.

(Smith, GW, and RE Reynolds. 1992. J. Wildlife
Manage. 56306-316.)
12
Empirical evaluations compensatory vs. additive
mortality
(Poysa H et al. 2004. Oikos 104612-615)
13
Empirical evaluations compensatory vs. additive
mortality

• Used three experimental manipulations to test the
  hypothesis of compensatory mortality in a
  Colorado mule deer population.

• Focused on survival of fawns and used
  radio-collared deer.

Bartmann, RM et al. 1992. Wildlife Monographs No.
121.
14
Experiment I Density reduction in field

• Hunting mortality was simulated by live trapping
  and removing 20 of the population on half of
  the Ridge study area in November-December.

• Estimated mortality rates of fawns on treatment
  and control areas until June.

• No spatial replication

15
Experiment II Controlled density in pastures

• Stocked three pastures with different deer
  densities in winter

• Estimated over-winter fawn mortality rates

• Simulated situation in which all pastures stocked
  as same high density (133 deer/km2) and
  then different harvest levels imposed (67, 33,
  0)

• No spatial replication

16
Experiment II Controlled density in pastures

• Fawn survival was related negatively to density.

17
Experiment III Predator removal

• Tested null hypothesis that decreased predation
  rate on mule deer fawns does not affect overall
  survival rate

• Predation rates decreased and starvation rates
  increased, but no change in overall fawn survival
  was detected.

• Concluded that results again supported
  compensatory mortality.

18
Maximum sustainable yield (MSY)

• Dominant concept in harvesting for many years in
  fisheries/wildlife.

• The largest average harvest that can be
  continuously taken from a population under
  existing environmental conditions without driving
  population toward extinction.

• MSY equals the maximum rate of recruitment, and
  it is obtained by depressing population to
  density at which the recruitment curve peaks
  (always below K)

19
MSY for logistic growth model
Peak recruitment
½ K

• Nu is a stable equilibrium (values gt Nu will
  decrease to Nu values between NL and Nu will
  increase to Nu)

• NL is an unstable equilibrium (population moves
  away from it, up or down)

20
Potential problems with simple MSY harvesting
strategy

• Assumes that managers know K and current
  population size so they can harvest exactly the
  number of individuals to maintain population at
  MSY.

• Populations are not deterministic. Environmental
  variation is common.

• Age- or stage-structure can be important.

21
Harvesting Strategies

• Modeled fluctuations of ptarmigan population in
  Sweden (30 yr data)

• One of most important game species in
  Fennoscandia
• -100K hunters harvest 300K-500K ptarmigan/yr in
  Norway

• Examined how different harvest strategies affect
  mean annual yields, and how uncertainties in
  population estimates affect choice of strategy.

Annes, S. et al. 2002. Ecological Applications
12281-290.0
22
Harvesting Strategies
1. Constant harvesting

• Provides stable yield
• Models suggest can drive population to extinction
  (especially populations with low growth rates and
  large stochastic fluctuations)

23
Harvesting Strategies
4. Threshold harvesting

• Requires estimate of population size (N) and
  threshold (c)
• Harvest estimated number of birds greater than
  threshold population
• Harvest N c for N gt c. Otherwise, no harvest.

24
Age-structured Harvest

• Hunters select certain ages
• Different ages contribute differently to
  population growth
• Example Elk in Yellowstone
• Hunters select middle-aged cows
• Wolves select young and old
• Population can support more predation from wolves
• (Wright et al. 2006)

25
Sex-structured Harvest

• Positive effects
• Hunter preference
• More breeding females ?
• Increased overall reproduction (polygynous
  system) ?
• Increased sustained yield
• (Caughley 1977)

26
Sex-structured Harvest

• Negative effects
• Example North American elk
• Normal mature malefemale ratio 25100
• Hunted population ratio 5100
• Conception and birth later and less compressed
• Juvenile mortality may increase 1 for each day
  past the median bird date
• Loss of predator swamping
• Timing with forage quality
• (Wisdom and Cook 2000)

27
Sex-structured Harvest

• Male-biased harvest can reduce population growth
  if immigrant males that replace a removed male
  kill young.

• Sexually selective infanticide hypothesis
  survival of cubs will be lower
  after resident bear is killed.

(Swenson et al. 1997. Nature 386450-451)
28
Harvest as a selective force

• Example African elephants
• Exploited for illegal ivory market
• Increased proportion of tuskless females
• Sex-linked, heritable trait
• (Jachmann et al. 1995)

29
Nuisance wildlife

• Public Safety
• 1.5 million deer collisions/yr
• 200 fatalities/yr
• Ecosystem Health
• Loss of floral diversity
• Disease transmission

30
Harvesting as a management tool

• Example Whitetail deer
• Allerton Park
• Unhealthy densities
• gt100 deer/ mi2
• Severe destruction to plant community
• High disease risk
• Initiated archery hunt to reduce herd

31
Adaptive Harvest Management (AHM)

• Formal evaluation of management options
• Continual monitoring and modeling
• Population status
• Habitat conditions
• Production
• Harvest levels
• Accounts for
• Structural uncertainty
• Partial observability
• Partial controllability

32
Summary

• More field experiments are required to understand
  better the mechanisms underlying the responses of
  wildlife populations to harvesting, especially
  compensatory mortality.
• Harvest regulations should err on the
  conservative side given the high degree of
  environmental stochasticity and uncertainty in
  parameter estimates.
• Harvest models should incorporate population
  structure
• Harvest can exert selective force in some
  situations
• Harvest can serve as an effective management tool
• Harvesting should be conducted as adaptive
  management within a flexible framework that
  allows for changes in regulations as new data are
  obtained.
• Harvest regulations always will reflect a mix of
  biological, social, and political considerations.