Transcript PVA

Viable population: population capable of selfmaintenance without continuing manipulation or
intervention
Minimum viable population (MVP)
• Survival probability of a population of a given size
for a designated period of time
Minimum viable population (MVP)
• Survival probability of a population of a given size
for a designated period of time
• specific to individual species and location; no
universal MVP exists
Minimum viable population (MVP)
Key issues:
• effect of chance events on
population persistence
• time frame for conservation
• probability level desired
Minimum viable population (MVP)
Key issues:
• effect of chance events on
population persistence
- scientific issue
• time frame for conservation
value
• probability level desired
judgments
Minimum viable population (MVP)
Example: red-cockaded woodpecker
• live in colonies (breeding pair plus helper offspring) requiring about
215 acres
• management plan across species range: 500 individuals needed in
each of 15 populations
Minimum viable population (MVP)
Example: red-cockaded woodpecker
• live in colonies (breeding pair plus helper offspring) requiring about
215 acres
• management plan across species range: 500 individuals needed in
each of 15 populations
• S. Carolina deme had mean time to extinction of 41.5 yrs, 72%
probability of extinction within 200 yrs
• annual addition of 3F + 2M for 10 years doubled projected mean time
to extinction; probability of extinction in 200 yrs reduced to 4%
Population viability analysis (PVA)
• A process to determine the probability that a
population of a given size will go extinct within a
given number of years
• Used to identify strategies for conservation
Population viability analysis - estimating MVP’s
• experimentation
+ powerful, species-specific tool
– ethical issues for endangered species
– takes too long: conservation plans needed within a
decade or two
Population viability analysis - estimating MVP’s
• simulation models
+ allow estimation of extinction probability (run 1,000
simulations, tally number of extinction events)
+ indicates which factors are most important in declines
– requires large amounts of data
– not generalizable - build anew for each species
Population viability analysis - estimating MVP’s
• simulation models
+ allow estimation of extinction probability (run 1,000
simulations, tally number of extinction events)
+ indicates which factors are most important in declines
– requires large amounts of data
– not generalizable - build anew for each species
• “off-the-shelf” programs available (VORTEX, GAPPS,
RAMAS metapopulation)
Basic procedure
Construct a computer simulation that projects the growth of the population into
the future.
1. Select population growth rate for each time step at random from a distribution
or set of possible growth rates. This will result in ‘good years’ and ‘bad
years’.
2. Repeat the project ion (e.g., 1000 x ) to estimate what the population is likely
to do on average.
Population viability analysis
• simulation models
– Bay checkerspot butterfly
current pop’n. trajectory
result of simulations (only 3 shown) at current
mean rate of population growth (r = 0.002)
Grizzly bears, the ‘original PVA’ – evaluating a single population
Bighorn sheep – evaluating a single population
Percentage of populations of bighorn sheep in North America that
persist over a 70-year period reduces with initial population size.
Koalas in Australia – comparing multiple populations
•Koalas are globally IUCN ‘least
concern’; secure in some areas,
but threatened in others in
Australia.
•National management strategy to
retain viable populations.
•Two populations modeled for
viability by Penn et al. (2000):
Oakey (declining) and Springsure
(secure).
Koala – Phascolarctos cinereus
•Populations modeled from 1970s
to 1990s to estimate extinction
probabilities (1000 iterations).
Variables in PVA
Values used as inputs for simulations of koala populations at Oakey (declining) and Springsure (secure),
Australia. Values in parentheses are SD due to environmental variation. The model procedure involved the
selection of values at random from the range. Catastrophes are assumed to occur with a certain probabilityin years when the model selects a catastrophe, reproduction and survival are reduced by the multipliers
shown.
Probability of
extinction
•Oakey=0.380 (380
out of 1000
iterations).
•Springsure=0.063
Observed koala population trends (diamonds) compared with
trajectories (triangles ± 1SD) predicted by 1000 iterations at (a)
Oakey and (b) Springsure, Australia.
Elephants – determining reserve size
• African elephants are IUCN Vulnerable.
• Armbruster and Lande (1992) modeled
population viability in 12 5-year age
classes through discrete 5-year time steps.
• Survivorship and reproductive rates
derived from Tsavo NP Kenya.
• Environmental stochasticity modeled as
drought based on Tsavo data.
African elephant – Loxodonta africana
• Habitat area and probability of extinction
examined in 1000-year simulations.
Results suggest that an area of 1300 km2 is
required to yield a 99% probability of
persistence for 1000 years.
Ocelots – assessing management options
Puerto Rican Parrot
(endangered)
Declined due to poaching (pet trade),
predation, cyclones, to N = ~13 birds
Puerto Rican Parrot
(endangered)
13-14 animals in wild; captive pop’n started
PVA: with 40 birds, 30% prob. of extinction
in 100 yrs
- primary risk: catastrophe (hurricanes)
Puerto Rican Parrot
(endangered)
13-14 animals in wild; captive pop’n started
PVA: 30% prob. of extinction in 100 yrs
primary risk: catastrophe (hurricanes)
Strategy: stockpile food
establish 5 populations
Puerto Rican Parrot
(endangered)
13-14 animals in wild; captive pop’n started
PVA: 30% prob. of extinction in 100 yrs
primary risk: catastrophe (hurricanes)
Outcome: hurricane devastated forest
habitat; captive population saved due to
food stockpile
Puerto Rican Parrot
(endangered)
13-14 animals in wild; captive pop’n started
PVA: 30% prob. of extinction in 100 yrs
primary risk: catastrophe (hurricanes)
Issues: population fragmentation, diseases
from captive birds
Florida panther:
PVA inputs:
population size
sex ratio
age distribution of each sex
age at first reproduction
maximum breeding age
% of each sex breeding each year
sex ratio at birth
mating strategy
number of offspring per year
probability of survival at each age
harvest
probability of catastrophic events
estimated carrying capacity
Florida panther:
declined due to habitat loss, poaching, road kills
evidence of inbreeding:
low fertility, sperm abnormalities, cowlicks, kinked tails,
cardiac defects, undescended testicles, high disease rate
1989 PVA:
at N < 50, predicted decline of 6-10%/year,
extinction in 25-40 yrs
- possible earlier extinction due to disease
Florida panther:
Outbred with sub-species from Texas - added 8 females in 1995
F1 hybrid kittens do not have cowlinks or kinked tails
Texas genes are now 15-29% of total
Road kills lower due to addition of culverts
PVA in 1999: extinction probability much
lower
PVAs – a work in progress. . .
Proportion of recovery plans written for species listed under the US Endangered Species Act that
present information regarding PVA.
PVA limitations
• Accuracy depends on quality of data.
• Lack of data on variables problematic.
Criteria for successful (accurate) PVAs
• Available data must be of sufficiently high quality to yield accurate descriptions
of all parameter distributions (e.g., population growth rates or rates of survival
and reproduction.
• Distributions of population parameters must hold fairly constant into the future.
PVA assumes that future population dynamics will be similar to those captured
in the data in hand, which reflect past trends.
Greater kudu - Tragelaphus strepsiceros
Mammalian PVAs
The predicted area required to support a given minimum
population size varies with
– the mammal’s feeding ecology (herbivore vs. carnivore)
– the mammal’s environment
• temperate vs. tropic (area required is larger for tropical species)
• high environmental variance requires more area
Mammalian PVAs
The predicted area required to support a given minimum
population size varies with
– the mammal’s feeding ecology (herbivore vs. carnivore)
– the mammal’s environment
• temperate vs. tropic (area required is larger for tropical species)
• high environmental variance requires more area
In general,
–
–
–
–
larger mammals require smaller populations, but larger habitat areas
carnivores require more habitat area than herbivores
populations with more variation in growth rate must be larger to persist.
densities tend to be lower in tropics (may be due to greater spp. diversity,
leading to greater number & intensity of biotic interactions)
Model predictions about species persistence in
the parks of the world
(Frankel & Soulé 1981 - Conservation and Evolution)
For largest mammalian carnivores (10-100 kg)
0-22% of parks should permit persistence for 100 yrs
no parks expected to permit persistence for 1000 yrs
For largest mammalian herbivores (100-1000 kg)
4-100% of parks should permit persistence for 100 yrs
0-22% of parks should permit persistence for 1000 yrs