Biodiversity - University of Windsor

Download Report

Transcript Biodiversity - University of Windsor

Biodiversity
Conservation Biology 55-437
Lecture 3
Feb. 26, 2010
Global Biodiversity: Patterns and Processes
Biodiversity is the diversity which exists in the
biological realm, either locally or over the globe.
Hierarchies of Diversity
Noss, RF. 1990. Con. Biol. 4:355-364.
Genetic Diversity
Genetic
Structure
Genetic diversity reflects the evolutionary history
of a population and how it will evolve in the
Genetic
future.
processes
The number of genes found in different species ranges
over orders of magnitude:
500 in bacteria to 20,000 in a mouse
• Among species: determine phylogenetic relationships
•Within populations: used to identify forces acting on
genetic variation
Genes
Population Diversity
• Demographics: fecundity, recruitment, mortality, etc.
• Distributions: relative abundance, density, etc.
• Population Structure: sex ratios, age distributions
Population
Structure
Includes genetic diversity
Genetic
diversity
Genetic and Population
Diversity
• Jeopardy question Dec. 15 2009:
• What North American Bird stands ~ 5 ft.
tall and had a population size of only 21
birds in 1940?
Diversity of Species
• Species are the fundamental units of
evolution.
• Focus of Legislation:
• CITIES - Convention on Trade in Endangered Species (Global)
• SARA - Species at Risk Act (Canada)
• ESA – Endangered Species Act (USA)
• Is focusing conservation legislation on this
level good or bad?
How Many Species are there?
The Geographical Distribution of Biodiversity
To quantify and describe the distribution of diversity
there are 2 common scales to measure diversity:
1.Richness (a count), as (single) point richness. It
includes no component of relative abundance.
2.As any of a number of measures that include relative
abundance:
alpha () diversity: a measure over a small homogenous
area
beta () diversity: rate of change of species composition
over a habitat gradient
gamma () diversity: changes over entire landscapes
Hierarchies of Diversity
 diversity
 diversity
 diversity
 =  / avg. 
There are a number of different measures of , β, and
γ diversity that incorporate relative abundance:
1.The Simpson (or dominance) Index. The
mathematical formula is:
D
1

s
i 1
pi
2
where i is the subscript identifying species and s is
the number of species in the sample. pi is the
proportion of total abundance represented by
species i.
2. The Shannon-Wiener (Information Theory)
Index.
It has been widely used for decades since Del
Shannon and Norbert Wiener invented the index
for code breaking during World War II. The
mathematical formula for diversity is:
H' 

s
i 1
pi log 2 pi
Relative abundance is assessed in part using
evenness, and is based on the ratio of the observed
diversity index to the one which would have been
found had all species been equally abundant.
Evenness (H'/Hmax) is also called equitability. This
measure was developed by Edith Pielou. The
formula for Hmax is:
H max  log 2 s
3. Brillouin’s Index
This index is similar to information theory, but
where information theory could use biomass or
another measure of relative abundance, Brillouin’s
explicitly uses the number of individuals.
Mathematically:
1
N!
H
log
N
N 1 ! N 2 ! N 3 !... N s !
4. Fisher’s 
This index arises from the mathematics of an
assumption that the abundances of species in a
community follow a log series distribution. That is
approximately the case for relatively low diversity
communities. The mathematics involves iteratively
fitting two parameters from a known number of
species and total number of individuals…
S    ln(1  X )
X
N
1 X
Geographic Patterns of Diversity - Plots of
Physical Variables
Geographical survey can be developed at two levels:
1. Classical division of the various biomes along
gradients in basic physical variables. The nicer the
climate is, the more diverse the community of
species should be.
An example: We would expect low diversity in polar
desert communities, and the diversity is very low.
There is little precipitation and a very low rate of
decomposition, so that nutrients are not readily
available and soils are poorly developed.
The classical division of biomes based on climate
was produced by Whitaker. It has one glaring
weakness: it does not separate the various forms of
grassland (steppe, savanna, grassland, tall grass and
short grass prairies) very well.
The various biome types have a reasonably well
established geographical distribution over the globe.
The map tells us where biomes are, not a global
diversity pattern. There is a general pattern of
declining species richness with increasing latitude.
This same pattern extends from whole communities
to species within most taxa. Here is the pattern for
bivalve mollusks:
And 2 versions of the broader taxonomic pattern in
the Americas…
And finally in a single smaller taxonomic group,
ants, in the U.S. Note that all the hot spots are located
in the southern half (concentric rings indicate higher
values ‘in the center’.
Ricketts et al. (1999) showed clear reductions in
species richness going northward from lower latitudes
(south Florida). The reductions were greatest for
vascular plants > birds > butterflies > trees > land
snails > mammals > reptiles > amphibians > beetles.
The general pattern is one described as latitudinal
gradients in diversity.
We can dismiss one hypothesis (as obvious) early.
There are more species at lower latitudes because
there are more habitat types. Why?
Adiabatic lapse means that at higher elevations in
tropical areas the cooler climates of temperate areas
are reproduced, and at extremely high elevations arctic
conditions may occur. The converse is impossible;
there is no means for tropical conditions to be
reproduced in temperate latitudes. So inevitably, the
overall habitat diversity of the tropics is greater than
that at higher latitudes.
Sky Islands
However, the real question is why there is a higher
within-habitat diversity at lower latitudes?
Pianka (1994) provided a set of hypotheses and
explanations for these patterns.
1.Evolutionary time - diversity should increase with
the age of a community. It assumes that temperate
and more extreme latitudes remain impoverished as
a result of the cycles of Pleistocene glaciation.
Evolutionary response to the restoration of
interglacial climates is still in progress.
There are problems with this hypothesis. Tropical
communities were affected by recent glaciations.
The cycles of Pleistocene glaciation are argued to be
one of the most important forces in explaining
tropical forest diversity.
During each cycle of glaciation, continuous bands of
tropical forest became fragmented. Species
differentiation occurred in each of these fragments,
potentially during each cycle, so that what began as a
single tropical forest species at the outset of the
Pleistocene could have become 8 different species (4
cycles of glaciation: 12 4 8) times the number
of isolated fragments, which number at least 6-8.
The other problem is that the hypothesis is founded
largely on a northern hemispheric view.
Because land area is smaller at temperate latitudes in
the southern hemisphere, there was little Pleistocene
glaciation south of the equator. Should temperate
communities in the south temperate zone be
considered as 'young' as those at similar latitudes
north of the equator? (They are about equally
impoverished.)
A separate issue is repeated cycles of mass extinction
and re-diversification.
On average, diversity has increased over the
geological time scale, but the increase has not been
smooth and uniform. There have been a number of
episodes of mass extinction in which a significant
fraction of living taxa have disappeared over fairly
short times.
The rate of diversification following each mass
extinction was much higher than at other times, in
each case due to the availability of resources and
niche space.
Here are diagrams of marine family diversity and
extinctions through the last 550 MY, with the mass
extinction events indicated by *s.
2.Ecological time. It may not be evolution which is
needed to re-diversify habitats at higher latitudes,
but just re-immigration of species displaced by
glaciation.
Many areas of the Northwest Territories have only
been exposed for around 4000 years, and plant
species (e.g. black spruce) are still recolonizing.
Graham et al. (1996) reported that glaciation has
profoundly affected North American mammal
distributions. During Pleistocene glaciation,
species like muskox and caribou extended down
into this area (and farther south).
3. Climatic stability. A stable climate is one which
changes little over time, both seasonally and from
year to year. A species living in a stable climate can
evolve specialized adaptations to the specific
climate. One which lives in an unstable or
unpredictable climate must have broad tolerance
limits, and, logically, broad niches. That leaves
niche space for fewer species.
4. Climatic predictability. If a climate is highly
predictable, the species can evolve life history
adaptations which reflect climatic cycles, for
example winter or drought dormancy.
5. Spatial heterogeneity. The more heterogeneous
the habitat, the more ways species can exploit it.
There are more possible niches. The number of bird
species increases with the foliage height diversity in
both North America and Australia
6. Productivity. With more resources there are likely
to be a greater number of individuals in the habitat.
Whatever the distribution of abundance among
species, a greater number of individuals logically
results in a greater number of species.
Same area
(representing
a number of
individuals) in
each band
There are important exceptions to productivitydiversity relationships. Estuarine areas, among the
most productive in the world, are very species-poor
when compared to other habitats of similar
productivity.
Conversely, some areas of restricted productivity are
far more diverse than their productivity suggests.
Example: plant diversity in the extreme southwest of
Australia, which includes an unlikely diversity of
Eucalyptus and Acacia species in small areas
Plant species diversity
Soil infertility may be a driver – low fertility favors
ability to exploit slightly different microhabitats.
Whatever the questions, there does seem to be a clear
relationship between productivity and diversity.
Ricklefs (with collaborators) demonstrated the
relationship using evapotranspiration as an indirect
measure of photosynthesis…
7. Stability of primary production. Extend the
arguments about climatic stability to stability in the
energy supply available to food chains and webs.
More species can be supported with a finer division
of resources if the amount of available resource is
predictable.
8. Competition. If competition is intense, then
selection produces populations which have
differentiated niches. Specialization which results
from competition leaves narrower niches and
greater diversity.
9. Disturbance. This is essentially the antithesis of
the competition hypothesis. Disturbances reduce
the intensity and effect of competition, and reduce
the diversity. In undisturbed communities
competitive dominants occupy most of the space in
the community. In very frequently disturbed
communities pioneer (weedy) species dominate.
However, intermediate frequency and/or intensity
of disturbance can clear space in a community, and
allow diversity to increase.
This idea is called the intermediate disturbance
hypothesis (Connell 1978).
10.Predation. Predation reduces the population size
of dominant species. That rarefaction reduces the
intensity of competition among prey, and can
permit the coexistence of species which would
otherwise suffer competitive exclusion.
Whether there is a latitudinal gradient to be
expected in predator effects is open to question.
Whether diversity can be affected locally is not in
question.
11.Species-area relationships. Rosenzweig (1992)
proposed that latitudinal gradients in diversity were
the result of a simple area relationship. Tropical
habitats immediately north and south of equator
abut one another, thus total tropical habitat is much
greater than for any other ecoclimatic zone. Larger
areas are assumed to result in higher speciation
rates and lower extinction rates, and thus higher
diversity.
Chown and Gaster (2000) criticized this hypothesis
with three lines of evidence:
a. there is no relationship between species’ range
size and habitat area available in the biome;
Species / Energy Relationships
For mammals, body mass is tightly linked
with density.
K. J. Gaston and T. M. Blackburn, Patterns and Process in
Macroecology. Blackwell Scientific, Oxford, UK, 2000.
b. there is no relationship between species’ range
size and speciation rate; and
c. there is general support (not conclusive however)
for the idea that extinction rate declines with
habitat area.
a, b and c should all be true to support Rosenzweig’s
hypothesis.
12.Evolutionary Speed. Higher temperatures in
tropics fosters an elevated speciation rate since
generation times are lower.
Many more hypotheses have been proposed, and no
single answer alone is likely to be correct. This is a
recent table (Willig 2003) of the many suggested
hypotheses:
Why biodiversity is important:
Tilman and Downing (1994) reported that primary
productivity of highly diverse grasslands was more
resistant to, and recovered more rapidly from,
drought than less diverse grasslands.
When we consider the potential impact of global
change, high biodiversity may be a protective factor.
Finally, a little supplementary information
about known causes of the mass extinctions:
a) Perhaps the leading explanation is the
comet (or asteroid) impact (or Alvarez)
hypothesis. Impact of a small comet or
asteroid would create a dust cloud far larger
than would be created by any known nuclear
weapon. Evidence exists that such events
occurred in the Cretaceous Period.
Dust in the atmosphere from these impacts would have
created the natural equivalent of a nuclear winter. The
Cretaceous mass extinction killed 16% of marine and
18% of land vertebrate families.
b) The Triassic mass extinction (~200
MYBP) may have been caused by massive
mid-Atlantic magma/ volcanic activity
that rifted Africa from South America. It
would have caused enormous global
warming.
The toll of this warming: 22% of marine families, and
an unclear number of terrestrial families
c. The cause of the Permian mass extinction is not
clear. It may have been caused by an asteroid
collision, or by vulcanism arising from such a
collision. It occurred ~251 MYBP. It was the most
devastating extinction, killing 95% of all species,
including 70% of terrestrial species of all kinds.
d. The Devonian mass extinction (364
MYBP) is unexplained. It resulted in the
loss of 22% of marine families and 57%
of marine genera.
e. An emerging hypothesis suggests that the end-ofOrdovician extinction (~440 MYBP), which wiped
out about 66% of species 440 million years ago, could
have been caused by ultraviolet radiation from the sun
after gamma rays destroyed the Earth's ozone layer.
It’s been suggested that a supernova exploded near the
Earth, destroying the chemistry of the atmosphere and
allowing the sun's ultraviolet rays to cook fragile,
unprotected life forms. These ideas were suggested in
2003 by Adrian Melott, a University of Kansas
astronomer.
Fossil records for the Ordovician extinction show an
abrupt disappearance of two-thirds of all species,
while other records show an ice age that lasted more
than a half million years started at the same time. Sea
level first fell with glaciation, then rose with glacial
melting. Melott said a gamma ray burst striking the
Earth would break up molecules in the stratosphere,
causing the formation of nitrous oxide and other
chemicals that would destroy the ozone layer and
shroud the planet in a brown smog.
The losses: 25% of marine families and 60% of
marine genera.
Managing Habitat: Wildlife Preserves
Discuss Wildlife Reserves, etc.
References –
Brown, J. and M.V. Lomolino 1998. Biogeography. Chapter 5.
Cameron, T. 2002. 2002: the year of the ‘diversity-ecosystem function’ debate. Trends
in Ecology & Evolution 17:495-496.
Chadwick-Furman, N.E. 1996. Reef coral diversity and global climate change. Global
Change Biology 2: 559-568.
Chown, S.L. and K.J. Gaston. 2000. Areas, cradles and museums: the latitudinal
gradient in species richness. Trends in Ecology & Evolution 15:311-315.
Connell, J.H. 1978. Diversity in tropical rain forests and coral reefs. Science 199:13021310.
Graham et al. 1996. Spatial response of mammals to late Quaternary environmental
fluctuations. Science 272:1601-1606.
Heywood, V.H. (ed.) 1995. Global Biodiversity Assessment. UNEP. Cambridge Univ.
Press, Cambridge.
Johnson, K.H. et al. 1996. Biodiversity and the productivity and stability of ecosystems.
Trends in Ecology & Evolution 11:372-377.
Kruger, F.J. and H.C. Taylor. 1979. Plant species diversity in Cape fynbos: Gamma and
delta diversity. Vegetatio 41:85-93.
Latham , R.E. and R.E. Ricklefs. 1993. Global patterns of tree species richness in moist
forests: energy-diversity theory does not account for variation in species richness.
Oikos 67:325-333.
Pianka, E.R. 1994. Evolutionary Ecology. 5th Ed. Harper & Row, N.Y.
Price, A.R.G. 2002. Simultaneous ‘hotspots’ and ‘coldspots’ of marine biodiversity and
implications for global conservation. Marine Ecology Progress Series 241:23-27.
Recher , H.F. 1969. Bird species diversity and habitat diversity in Australia and North
America. American Naturalist 103:75-80.
Rice B. and M. Westoby. 1983. Species richness at tenth-hectare scale in Australian
vegetation compared to other continents. Vegetatio 52:129-140.
Ricketts, T.H., E. Dinerstein, D.M. Olson and C. Loucks. 1999. Who’s where in North
America? Bioscience 49: 369-381.
Smith, F.D.M. et al. 1993. How much do we know about the current extinction
rate? Trends in Ecology & Evolution 8:375-378.
Willig, M., D. Kaufman, and R. Stevens. 2003. Latitudinal gradients of biodiversity:
Pattern, process, scale, and synthesis. Annual Review of Ecology and Systematics 34:
273-309.
Tilman, D., and J.A. Downing. 1994. Biodiversity and stability in grasslands. Nature
367:363-366.