Transcript M. pinetorum

Species Diversity and
Community Stability
Some initial observations


When we sample species in a
community, we usually find a few
species that are very common, while
many species are rare.
In our trapping efforts this semester in
Ecology and Mammalogy, we have
trapped primarily B. carolinensis,
followed by P. leucopus, M. pinetorum,
and O. nuttalli.
Initial observations.

In fact, we usually find the data follow a
logarithmic series:
x,
x x x
2
2
3
,
3
,
4
4

Initial observation.

Here, x = number of species in the
total catch represented by one
individual, x2 = number of species in
the total catch represented by two
individuals, and so on. Then, the
number of species in a sample is S =
xn
x,
x
x x
2
2
,
3
3
,  ,
x
n
n
Relative abundance of butterflies
in Rothamsted, England, in 1935.
Initial observations

What does this tell us about the
structure of communities?
Diversity Indices


We need some index to evaluate the
diversity of species in a community.
A common, and reasonable index is the
Shannon-Wiener Diversity Index.
S
H '   pi log e pi
i 1
Diversity Indices


Here, pi = proportion of the ith species in
the total sample of S species.
This index has the pleasing property,
that communities with uneven
abundances of species have lower
diversity.
Diversity Indices

Compute H’ for each of the following:
– Community 1 with 90 individuals of species
A and 10 individuals of species B.
– Community 2 with 50 of species A and 50
of species B.
– Community 3 with 80 of species A, 10 of
species B, and 10 of species C.
– Community 4 with 33.3 of species A, 33.3
of species B, and 33.3 of species C.
What do you get?





Community 1) H’ = 0.33
Community 2) H’ = 0.69
Community 3) H’ = 0.70
Community 4) H’ = 1.10
These results are exactly what we
would expect intuitively.
Evenness

We can estimate the evenness of the
community by using J’, where
J' H '
H 'max
Evenness

Here, H’max is the maximum possible
diversity, assuming all species in the
community have equal representation.

Of course, these estimates are valid
only within the context of any given
study, and are difficult to compare
across studies. Do you know why?
What if you need to compare
indices across studies?

The best bet is to rely on Species
Richness. This is simply the number of
species observed.
Gradients of Species Diversity

In a very general way, we know that the
tropics contain more species than the
temperate zones. For example,
– there are more than 1000 species of fish in
the Amazon, 456 in Central America, and
only 172 in the Great Lakes.
– There are 7 ant species in Alaska, 73 in
Iowa, 101 in Cuba, 134 in Trinidad, and
222 in Brazil.
Gradients of Species Diversity

There are examples where the pattern
is opposite of what we expect:
– Sandpipers
– Aphids

Overall however, the patterns appear to
be clear.
Diversity of mammals
Isoclines
of
mammal
diversity
in North
America
Isoclines
of avian
diversity
in North
America
How do we explain these
patterns?

Time Hypothesis

Spatial Heterogeneity Hypothesis

Competition Hypothesis

Predation Hypothesis
How do we explain these
patterns?



Climatic Stability Hypothesis
Productivity Hypothesis
Area Hypothesis: in larger areas, the
chances of isolation between
populations increase, with
corresponding increases in the chances
of speciation.
How do we explain these
patterns?


Animal Pollinators Hypothesis: in the
tropics and other humid parts of the
world, winds are less frequent and of
lower intensity than in temperate
regions. This effect is accentuated by
dense vegetation cover. Therefore,
most plants are pollinted by animals.
Can you think of others?
Community Stability



This is the ability of a community to
resist change following a disturbance
(=community resistance), or
the ability of a community to return to its
original configuration after a
perturbation(=community resilience).
Here it is worthwhile thinking about
equilibria from the Lotka-Volterra
analyses.
Community Stability



Deserts have high resistance.
Estuaries have low resistance, but high
resilience.
Does diversity cause stability?
– Laboratory experiments by Gause
confirmed the difficulty of achieving
numerical stability in simple systems.
Community Stability
– Small, faunistically simple islands are
much more vulnerable to invading species
than are continents.
– Outbreaks of pests are often found on
cultivated land or land disturbed by
Humans: both of which contain few
species.
– Tropical rain forests do not have insect
outbreaks like those common in temperate
forests.
Community Stability
– Pesticides have caused pest outbreaks by
the elimination of predators and parasites
from the insect community of crop plants.
– In a review of 40 food webs, the complexity
of food webs in stable communities has
been found to be greater than the
complexity of food webs in fluctuating
environments.
Community Stability

If diversity is equated with stability, then
stability =
S
H '   pi log e pi
i 1
Community Stability


In a food web with 4 links (1 predator
and 4 prey), each link caries 0.25 of the
total energy in the food web, and
stability = -(4 x 0.25 x log(0.25)) = 1.38.
Adding another predator that eats all the
prey doubles the number of links to 8,
and stability = 2.08.
What are the stability implications of these webs?
Community Stability


We can also get a given stability by
having a large number of species, each
with a restricted diet (specialists), or a
smaller number of species each with a
broader diet (generalists).
Maximum stability occurs when there
are m species and m trophic levels, with
each trophic level containing 1 species.
Community Stability



Does this make sense?
Restricted diets lower stability in general, but
in practice specializations may be essential
for efficient exploitation of prey.
In arctic systems with few species, it is
difficult to have a specialized diet, and
species are generalists (=greater stability),
but there are few species and thus
populations fluctuate considerably.
Community Stability

In the tropics, with many species,
stability can be achieved with restricted
diets, and species specialize, feeding
on only 1 or 2 trophic levels.
Is there convincing evidence that
diverse communities are more
stable than simple ones?


1: fluctuations of microtine rodents are
as great in simple arctic communities as
they are in complex temperate
communities.
2: Some field data suggests tropic
stability is a myth (Robin Andrews).
Is there convincing evidence
that diverse communities are
more stable than simple ones?


3: Rain Forests seem particularly
susceptible to human perturbations.
4: Agricultural systems may suffer from
outbreaks not because of their
simplicity, but because their
components have no co-evolutionary
history.
Two alternative views:

Equilibrium Hypothesis
– Local population sizes fluctuate little from
equilibrium values, which are determined
by predation, competition, and parasitism.
Communities are stable and perturbations
are ‘damped’ out.
Two alternative views:

Non-equilibrium Hypothesis
– Species composition is constantly
changing, and never in balance. Stability
is elusive, and persistence and resilience
are key measures of community behavior.
– Key mechanism for this hypothesis is the
‘intermediate disturbance hypothesis’
Community Change

Succession
– If we were to burn the I.R. Kelso sanctuary
and then leave it alone, we could predict
fairly accurately what would happen.
• In the first few years it would be covered by
weeds and grasses.
• Shrubs would be established.
• Maple seedlings and other pioneer species
would be establsihed.
• Finally, after 30 or 40 years it would look like a
young Oak-Hickory forest.
Community Change


After 300 years, it would be a climax
community – an old growth forest with
little understory.
If we then burned it again, it would
repeat the sequence. However, once in
the climax, it will stay there unless
perturbed.
Community Change

The same pattern can be seen in the
coastal habitats of California, the Mesas
of New Mexico, the rocky intertidal, or
even hot-spring algal communities in
Yellowstone.

Why does this happen?
Rate of
ice
recession
in
Glacier
Bay,
Alaska
Community Change


Glacier recession results in significant
disturbance, with the newly exposed
habitats undergoing successional
change.
But, do the communities ever reach the
climax stage? There are many
examples where environmental
perturbations are frequent and prevent
attainment of the climax condition.
Community Change

There are 3 models for succession:
– Facilitation model
– Inhibition model
– Tolerance model
Facilitation model: each species makes the
environment more suitable for the next.
Inhibition model: Initial colonists tend to prevent
subsequent colonization by other species.
Succession depends on chance events (who invades
first). Succession proceeds as colonists die, but it is
not in an orderly or predictable fashion
Tolerance model: Any species can start the
succession, but the eventual climax is reached in a
somewhat orderly fashion.
Succession

Succession can be modified by a
number of factors:
– Stochastic events
– Life history
– Facilitative events
– Competition
– Herbivory
Influence of
succession and
environmental
severity on major
successional
processes that
determine change
in species
composition
during
colonization (C),
maturation (M), or
senescence (S).
Succession

What does all this mean?
– Succession is a complex process,
influenced by many factors.
Percent
vegetative
cover vs.
field age
and
nitrogen
conc. for a:
intruduced
plants, b:
non-prairie
natives, c:
true prairie
natives.
Island Biogeography


Studies of succession have benefited
greatly from studies of island
recolonization: Krakatau in 1883, Mt. St.
Helens in 1980.
When we study how ‘islands’ are
recolonized, we begin to understand a
great deal.
Island Biogeography

Area Effects
– What is the likelihood that an area will be
colonized? It depends on distance from
source pool, but also on ‘size of the target.’
– Also, a small habitat is unlikely to support
as many different types of colonists as a
big area.
Island Biogeography

This can be expressed as the following,
where S = number of species, c is a
constant measuring number of species
per unit area, A = area, and z = a
constant measuring the slope of the line
relating S and A.
S  cA
z
Island Biogeography

Remarkable, for a wide range of
species and island situations, z tends to
be about 0.3 (amphibians and reptiles of
the west Indies, beetles in the West
Indies, Ants in Melanesia, Vertebrates
in Lake Michigan, and plants on the
Galapagos.
Amphibians and Reptiles of the Antilles.
Flowering plants in England.
Insects of British trees: open dots are introduced
species.
North American Birds
Island Biogeography


Actual parameter values depend on
whether a true island is being
considered, size of the island relative to
number of possible colonists, and
colonizing ability of species.
The pattern also holds for nontraditional islands like mountain-tops in
the Great Basin.
Boreal birds
and mammals
in the Great
Basin.
Island Biogeography


Can we use these ideas to build a
model of island diversity?
This work was done by MacArthur and
Wilson back in the 1970’s, and
constitutes some of the most groundbreaking ecological work ever. Now of
course, it seems intuitively obvious. I
wonder why it was not obvious earlier?