Transcript Geographical Ecology

Biodiversity: An Equilibrium Theory
In considering the latitudinal gradient in diversity, some
factors increased local diversity (habitat heterogeneity and
structural complexity), some decreased local diversity
(competition), and some could act in either direction
(predation).
Like population growth models with birth rate balancing
death rate at equilibrium, there could be a similar model for
local diversity, with some processes adding new species and
others causing local extinctions of species.
That equilibrium theory was developed in 1967 in a seminal
book (The Theory of Island Biogeography) by Robert
MacArthur (of the warblers) and E.O. Wilson (of the ants).
They proposed that many communities occur in islands of
habitat, surrounded by inhospitable habitats where the
component species could not live. Such a situation is clearly
the case with oceanic islands.
They proposed that the equilibrium diversity is determined by
the balance between immigration of new species onto the
island and extinction of species already established there.
Here is the simple model:
Both immigration and extinction are curves. Why?
The immigrants must arise from a mainland (source) pool
that includes all species resident there. Some are very good
dispersers, others quite poor. The good dispersers likely
arrive first at a rate that reflects their dispersal ability.
After they’ve reached the island, what is left is poorer
dispersers, and their rate of arrival will be even lower than
simply proportional to the number of species left to
immigrate because of their lesser dispersal ability. Thus the
concave curve.
For extinction we begin with a simple proportionality that
says extinction is more likely when there are more species
that could go extinct. That would produce a straight line.
However, with more species the probability of some being
close competitors with others on the island increases, and
the probability of species having smaller population sizes
also increases. With both, the probability of extinction
occurring increases. Thus extinction rises faster than
simple proportionality, and is a curve.
What other factors influence these curves?
1. Distance of the island from a source of immigrants. The
rate of immigration will be lower onto more distant islands.
2. The area of the island. Population size will logically be
smaller on smaller islands, so that the extinction curve will
rise more rapidly.
Here are the curves:
Immigration from
near and far islands
Extinction on large
and small islands
The theory was tested in numerous studies from its initial
publication until around 1990, when metapopulation concepts
began to replace the island model.
Among the first tests were Jared Diamond’s measurement of
bird species diversity on islands surrounding New Guinea. He
found the predicted relationships of species numbers with
distance from
New Guinea
and with
island area…
The basic equation for the equilibrium species numbers on
islands includes only a few terms:
S  cA
z
S is the equilibrium number of species
c is a constant that reflects the dispersal capability of the
group of species under study (birds obviously have a
higher c than worms)
A is the area of an island
z is a constant that measures how rapidly the number of
species increases with area
But we know that there are other influences…
Diamond quantified the simple distance and area
relationships, and recognized that habitat heterogeneity on
islands might also be important. The full model equation
he developed included:
1. Island area – the basic model - here he found a power
function best fit his data. This equation was:
S = 15.1 A0.22
2. Then he added distance to the model…
S = 15.1 e-D/3750 A0.22
3. Finally, he added island elevation – elevation is an
indicator of spatial heterogeneity. For every 1000m of
maximum elevation on an island, bird species diversity
increased 2.7% on average.
S = 15.1 (1 + 0.027L/1000) e-D/3750 A0.22
Complicated though that equation is, it accounts for almost
90% of the observed bird species diversity among those
islands.
A second test was not passive observation, but a
manipulative experiment. It was Dan Simberloff’s Ph.D.
research.
He selected mangrove mangles (small islands of mangrove)
at different distances from the ‘mainland’ (the Florida
coastline), and of differing size. He collected the insects
from them to establish a baseline diversity. Some he kept as
controls; others he covered with plastic sheet and fumigated
with insecticide. He then followed the re-colonization of the
‘islands’ and the recovery of diversity on them. Here is how
he defaunated ‘islands’:
And here is what he found, measuring recolonization after
islands had been cleared of insects by fumigation:
E2 was the nearest island. It returned to a diversity of 43
species within about 1 year.
E1 was the farthest island. It had 24 species before
defaunation, and recovered to ~20 species over the two
years. Nearer islands recovered faster than farther ones.
Another important thing to recognize about the island model:
The equilibrium is a dynamic one. Over time there is
turnover, the replacement of one species by another in a list
of species that remains relatively constant in size.
Also note the broad applicability of the model; it applies to a
broad spectrum of situations where islands of suitable habitat
are separated in a matrix of unsuitable habitat. That isn’t only
true for oceanic islands.
What about mountains? Forests? Lakes?
The vegetation in zones along the slopes of mountains differs
a lot with elevation. A plant (or an animal associated with it)
from zone 2 is quite isolated by unsuitable habitat between
the mountains from the nearest other area with zone 2-type
vegetation.
Jim Brown made predictions about mammal diversity on
mountaintops, and how much might be lost through climate
changes that caused vegetation zones to shift in elevation.
The extinctions he predicts are severe.
For an arboreal, forest-dwelling bird, the grassland habitat
between patches of forest is unsuitable.
There are some remarkable examples of birds unwilling to
cross fairly narrow separations. No birds have been observed
to cross the 100-200m separating Barrow Colorado Island in
Panama from the Mainland.
For aquatic species living in lakes, the terrestrial
environments between lakes are clearly inhospitable.
Studies have shown the same distance and area
relationships in the diversity of species like snails in the
lakes of New York State.
Review of island biogeography
The features of islands that most influence diversity are
island area and distance from source(s) of species.
The simple rule for the effect of area is that a 10-fold increase
in area will double the number of species. This rule is usually
converted into a simple equation:
log S = log C + x log A
On a log-log plot, this equation shows you will see a straight
line.
x is the slope of the line
log S
C is the Y-intercept
log A
The observed values of x fall into a small range, from ~0.2
to ~0.33.
We might expect the slope to be different on ‘mainlands’ and
on islands. Mainlands will support more species, but add
fewer as you look at a larger area (x is lower). This results
from transients being counted from smaller areas on
mainland, thus not being added as area increases.
Mainland
Number of
Species
Island
Z = 0.12 - 0.17
Z = 0.24 - 0.34
Area (km2)
As islands increase in size, they become more like the mainland
That is the relationship. Why does area increase diversity?
a) Increased habitat diversity on larger islands support more
species
b) Larger populations of each species lowers chances of
extinction
c) There may be sufficient area (and thus population size) to
support (feed) an additional, higher trophic level
Distance reduces island species diversity. Why?
a) Differences in dispersal capacity makes it less likely that
poorer dispersers can reach distant islands. This is evident
even among good dispersers like birds with distance across
the Pacific from Australia and New Guinea as sources…
Birds
Australia Solomons Fiji Samoa Tonga Society
Pelicans
Quails,Owls
Crows
Hawks
Ducks
Pigeons
Starlings
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Samoa
New Guinea
Solomons
Fiji
Society Isl.
Tonga
The notion works in many different settings. For mosquitos
released in an urban area…
From Wolfenbarger 1975
Number
of
Mosquitos
Distance from release point
Immigration and Extinction are the processes that determine
island diversity. How do they occur, what constitutes an
immigrant?
1. Immigration
Immigration has two components:
dispersal to the island, which may be:
active (flight, walking) or
passive (carried on the wind, rafting)
establishing a breeding population on the island.
2. Extinction
The causes of extinction include:
random events (environmental uncertainty,
demographic uncertainties like the Allee effect),
loss of genetic diversity, or
deterministic events like predation and competition.
These events are more likely to cause extinction of small
populations than large ones.
Humans are responsible for a large fraction of
extinctions occurring today and in the recent past.
Let’s consider evidence of what we’ve done…
In the last century there have been 42 extinctions of
mammalian species; all have been on islands.
There have been 53 extinctions of bird species; 50 have
occurred on islands. Approximately 93% of bird species
extinctions between 1600 and 1980 were island occurrences.
In some cases we can tie those extinctions closely to
humans. Steadman (1995) documented the loss of bird
species on tropical islands following colonization of those
islands by humans. He used a combination of fossil records
and archaeological information on the islands.
The islands I used to show distance effects were among
those he studied.
Samoa, Tonga and the Society Islands were colonized by
Polynesians about 3500 years ago.
By 1000 years ago, human populations were widespread on
Pacific Islands in this area.
Taken in total, he estimated that 2000 species or subspecies
of birds were lost from Pacific islands over the 3500 years
since human colonization. That loss represents about 20% of
current global bird species diversity.
It was not always direct interaction with humans; frequently
it was inexperience with species carried along – dogs and
rats particularly.
Here’s details of the data for one island group, the Kingdom
of Tonga…
Period
Pre-human (>3000 years ago)
Bird Diversity
28 species
At the time of settlement (~1500 years ago) 21 species
200 years ago
10 species
Today
9 species
The largest percentage of losses occurred in conjunction with
or after the arrival of Polynesian people.
Back to the main notion – extinction and is causes…
Extinction increases as the diversity of species on an island
increases because:
1. there are more species to go extinct
2. Population sizes decrease as the number of species
increases. Smaller populations have a higher
likelihood of extinction.
3. Species interactions (competition, predation) may
hasten extinctions. They are more likely to occur as
the number of species present increases.
From the graphical model, we also know that extinction rate
at any diversity is higher on a smaller island.
There is evidence for that…
Stuart Pimm (1991) found that bird extinction rates on
European islands were strongly related to island area…
Extinction rate
(species/yr)
1
100
Area (km2)
10000
One of the implications of the model is that if ‘island’ area
were somehow reduced, there would be too many species for
the predicted equilibrium; species should go locally extinct.
Here is one neat ‘experiment’ that demonstrates this
phenomenon, called relaxation…
Barro Colorado Island was isolated by forming Lake Gatun
in building the Panama Canal. It had been a hilltop. The
‘island’ was isolated in 1914. It is only about 100m from the
shore of Lake Gatun, but the tropical birds will not cross this
barrier. Between 1914 and 1974, Barro Colorado lost 15
species resident at isolation. Slowly, but inexorably, it will
lose more.
Other aspects of history also play a role in where we find
species. This is a sneaky way to begin in Chapter 24…
First, an example of strange island populations. During the
cycles of glaciation, huge amounts of water were tied up in
the glaciers, and as a result, sea level dropped. In southeast
Asia, both among the Philippines and among islands around
Java, land bridges developed until sea level again rose. Some
islands became linked to the Asian mainland. That explains
how a large mammal like the rhinoceros (not noted as a
swimmer) is found on Java and Sumatra. Island
biogeography (relaxation) partially explains why their
number is declining, and these subspecies are considered
endangered.
Rhinoceros sondaicus
distributed among islands
of Malaysia.
50 left in Java
15 left in S. Vietnam
There is one last idea coming from island biogeography that
was initially hard to accept – that the equilibrium is
dynamic. The model and theory both indicate that there
should be turnover – replacement of species on an island
without a change in the total size of the list.
This was documented by Diamond, when he reviewed old
data from the 9 Channel Islands off the southern California
coast and compared them to new species lists he and Power
collected.
The first species lists came from Howell (1917). Diamond’s
survey came in 1968. Here’s a map showing you most of the
islands…
The 9 islands range from 3 –
>200 km2. They range from 5 –
40 km from the nearest source
of bird colonists.
We can detect extinctions if a
bird there in 1917 was not there
in 1968.
We can detect immigrations if a
bird found in 1968 was not
there in 1917.
How do the numbers compare?
Island
Los Coronados
San Nicholas
San Clemente
Santa Catalina
Santa Barbara
San Miguel
Santa Rosa
Santa Cruz
Anacapa
1917 1968 Extinctions Immigrations
11
11
4
4
11
11
6
6
28
24
9
5
30
34
6
10
10
6
7
3
11
15
4
8
14
25
1
12
36
37
6
7
15
14
5
4
There were many immigrations and many extinctions, but the
diversity on most islands remained relatively constant, since
rates of immigration and extinction were very similar.
The theory of island biogeography (and more recently ideas
drawn from metapopulation biology) have been applied to
conservation of species.
There are obvious reasons: human activities fragment
formerly continuous habitats, creating islands within which
the species must now live; we introduce exotic species that
interact with natives and increase their extinction rates since
there are fewer, if any, refuges; and we, as a species, poach
and hunt species, reducing their population sizes.
We, therefore, need to design areas within which species are
protected. They will be islands. How should we design
systems of such islands to maximize the chances of
persistence of species being conserved?
Design of Nature Reserves
Better
Worse
Design of Nature Reserves
Better
Worse
The suggestion that corridors are valuable to permit
movement of island populations among reserves has proven
more controversial than you might expect.
To preserve a sufficient corridor width to ‘protect’ a
migrating species is problematic. If wolves move 30-40 km a
day, does that mean a corridor to permit their movement
from one protected area to another needs to be 40km wide?
Wolves may not be the best example. Smaller species need
far narrower corridors, but their predators may not have easy
access in the preserve. Is access equally difficult when they
move along a narrow corridor? No!
Loss and fragmentation of tropical forest habitat has been
particularly severe. Here are some careful estimates of
percentage of tropical forest lost in a few countries of Africa
and southeast Asia:
Country
Kenya
Madagascar
Malaysia
Thailand
Bangladesh
% lost
48
75
41
74
94
The development of pasture land in the tropical forest area of
Brazil has had parallel impact. But Brazil has at least
attempted to protect areas…
To develop new pasture area, Brazil now requires that a
significant percent of total area owned be preserved as
undamaged tropical forest. In addition, there are separate
reserves for indigenous peoples and as national parks.
Costa Rica has also legislated protection of large portions of
remaining tropical forest as reserves. In Costa Rica,
exploitation of those protected areas now takes the form of
ecotourism.
Finally, what is now a classic example of the impact of
exotic species. In moving military cargo among Pacific
islands after WWII, the brown tree snake, Boiga irregularis,
was introduced onto Guam. It is arboreal and has caused
numerous extinctions of endemic birds on Guam (as well as
numerous power outages when it suicidally shorts power
lines).
What follows is a map of Guam. The arrows indicate known
expansion of the range of the snake over Guam. For a
number of sites, the number of forest birds in different years.
There were once (1940s) 10 species of forest birds on Guam.
At each location, once the brown tree snake had reached the
area the number of forest birds declined. There are now only
3 species left: 50 starlings at one site, 200-300 swiftlets at
one site, and 20 crows at a third site.
Boiga won’t starve like a predator that has driven its prey
extinct. Humankind also carried rats to Guam; Boiga likes
rats, too.
The current concern is that the brown tree snake may invade
the Hawaiian Islands, moved there by human traffic. It can
survive for longer than the flight time from Guam to Hawaii.
Dead brown tree snakes have already been found in cargo at
Hawaii.