Conservation At The Population And Species Level (Lec 6)

Front 1

Is species conservation enough?

Back 1

focuses on single threatened species

crisis approach to conservation

should be complemented by bigger-picture approaches

Front 2

Primary IUCN criteria for listing

Back 2

range size (small is worse)

population abundance (small is worse)

population trend (declining is worse)

ongoing conservation efforts

used to determine extinction risk

Front 3

Population density affected by

Back 3

immigration and emigration

births and deaths

Front 4

What is a model?

Back 4

a simplified and abstracted representation of a system

allows hypothesis testing and generalization

Front 5

GEP box on models --->

Back 5

all models are wrong but some are useful

Front 6

Exponential growth model

Back 6

N(t+1) = N(t) + N(t)*r

N(t) = N(0) * lambda(t)

Front 7

Geometric rate of increase

Back 7

fundamental parameter in population ecology is the finite rate of increase of lambda

Front 8

lambda (in the exponential growth model)

Back 8

describes the proportional increase (or decrease) in a population per time step

Front 9

Did lambda apply to humans?

Back 9

yes

Front 10

lambda < 1 means

Back 10

decline and eventual extinction

Front 11

lambda = 1 means

Back 11

stable population

Front 12

lambda > 1 means

Back 12

the population is growing

Front 13

Why models are important for conservation

Back 13

assessment of extinction risk

summarizes what is known about a population

allows projections into the future

assumptions are made explicit

model can be refined as amore data become available

model parameters can be compared among species and populations

Front 14

Example of rates of decline

Back 14

Northwest Atlantic sharks

Front 15

example of re-introduction

Back 15

muskox

reintroduced to Nunivak Island, AK

Census data collected to build a model in order to manage the population

growth rates averages at lambda = 1.148

Front 16

Next step for the models?

Back 16

include real-world variability and uncertainty

Front 17

So far models have assumed what?

Back 17

that species grow in a simple, predictable fashion

assumed that growth parameters are fixed

Front 18

deterministic model

Back 18

N(t+1) = N(t) * lambda

when growth parameters are fixed

Front 19

stochastic model

Back 19

N(t+1) = N(t) * lambda(t)

takes into account that parameters vary widely

more realistic

Front 20

Causes of variability

Back 20

1. natural - population variability and environmental variability

2. sources of uncertainty - parameter uncertainty

- model uncertainty

- ignorance and surprises

Front 21

What does stochastic models include

Back 21

both variability and uncertainty

Front 22

Persistence vs. initial population

Back 22

can calculate probability of extinction (using same model) within 100 time steps

Front 23

Environmental variability

sources of environmental variation

Back 23

temporal

spatial

extreme

Front 24

temporal variation

Back 24

light, temperature, rainfall, predation, food supply, etc.

Front 25

spatial variation

Back 25

along depth gradients, across continents

Front 26

extreme variation

Back 26

events or catastrophes - can harm or benefit

Front 27

How do we incorporate environmental stochasticity

Back 27

assume that lambda caries over time through a statistical distribution

Front 28

Muskox model

where lambda = 1.148 and k = 500

density dependence: Ricker

Back 28

type: deterministic

probability of extinction = 0

Front 29

Muskox model

lambda = 1.148 and k = 500

Back 29

type: stochastic (N=25) using demographic stochasticity

N is how many many times the model is run

the line is the mean and the data includes error as well as the highest and lowest value

Front 30

same model but adding a standard deviation in lambda of 10%

Back 30

adds a lot of variability to the data

Front 31

add a standard deviation in lambda of 30%

Back 31

trajectory summary means that the probability of extinction within 50 years is about 10%

Front 32

Problem: parameters are always estimated, never known

Back 32

solution: use larger sample size, get additional information to reduce uncertainty

Front 33

problem: models are a simplification of the real world

Back 33

solution: compare different models and see how their behaviour varies

Front 34

Overall solution to redce uncertainty

Back 34

find a good compromise between simplification and realism

Front 35

Allee effects

Back 35

correlation between population size or density and the mean individual fitness of a population or species

Front 36

allee effects in great auk

Back 36

overhunting by museum collectors

Front 37

allee effects in Napolean wrasse

Back 37

overfishing increased price

Front 38

allee effects in passenger pigeon

Back 38

failure of large breeding colonies

Front 39

urchin

Back 39

critical density needed for urchin fertilization succes

Front 40

populations of most concern

Back 40

populations in rapid decline

Front 41

stochastic models are used for....

Back 41

to assess population variability and uncertainty