Part1

Document Sample
Part1 Powered By Docstoc
					Conservation Biology (Biol 4350/5350)
Fall 2012

Chapter 1: What is Conservation Biology?


(A fairly brief overview)
Almost all ecosystems on Earth have been
impacted by humans to some extent; rate
and degree is increasing exponentially
with growth of the human population
(about 6 billion people currently).

Climate change, pollution, biogeochemical
cycles, extinctions, habitat destruction and
change, loss or alteration of genetic
variation...

Affects humans directly and indirectly.
1.1 Estimated global human population size from
the last Ice Age to the present
1.2 United Nations projections for human population growth to 2050
1.3 Number of global hectares per person needed to support current lifestyles
1.4 Map of the human footprint
BUT:

Human growth rate has slowed in many
countries.

Much depends on how people act and
how much they consume, not just how
many there are -- and consumption is
much lower in 3rd World countries; this
can change in developed countries too.
And:

Highest birth rates are in places where
many family members are needed to
achieve success at low-skill tasks.

Education and targeted economic
development can reduce the incentive to
produce large families.
Try to achieve a balance between human
needs and maintenance of biodiversity
through sustainable development.

     Conservation biology: Many ideas are
old, but really became a science in the
1980s: ties together areas of biology such
as wildlife and habitat management, to
population genetics, to evolution, to
ecological modeling etc.
     Integrates with chemistry, geology,
anthropology, sociology, economics...
Elements unique to the “new”
  conservation biology:

1) Spans the previous gap between pure
   and applied research; management
   plans consider genetic, theoretical, and
   many other kinds of data (not just
   counting trees, deer, etc.).

Now, conservation biology is firmly
  established as rooted in real academic
  research.
2) Earlier approaches were largely human-
  oriented (anthropocentric) and
  utilitarian: focused on maintaining
  species that people use or like (timber,
  fisheries, flashy species etc.), and
  “useful” components of ecosystems
  (e.g., water).

Now much more concern about diversity
  of a wide range of species (which may
  in turn help preserve the ones that are
  directly useful to humans).
And -- growing recognition that biodiversity
  itself has inherent value; healthy
  ecosystems are the “support system” of
  the planet.

(Also, wider view of “intrinsic value” --
  biodiversity is valuable regardless of its
  direct benefits to humans -- we’ll
  discuss this more later).
3) Realization of the importance of
  integrating conservation strategies with
  work of non-scientists: economics,
  politics, social sciences, urban planning,
  etc.

People have to cooperate, establish viable
  alternatives, and understand the
  benefits of preserving biodiversity.
In the broadest sense, conservation
   biology tries not only to maintain diverse
   species, but also genetic diversity and
   genetic “integrity”: protect populations
   and gene flow.

And, keep entire ecosystems functioning
  normally while recognizing that change
  (ecological, evolutionary) does occur
  even without human intervention.
Conservation biology, from a scientific
 perspective, is an attempt to maintain
 normal evolutionary processes within
 normally functioning ecological settings.
Even in ancient Greece (for example),
  massive habitat destruction was in
  progress, and was recognized.

Huge areas of S. Europe, Mediterranean,
  and SW Asia were once called “the land
  of perpetual shade”; forests were
  destroyed to build ships etc.; now
  barren and/or desert.
The degree of human impact has
  depended partly on population density
  and partly how long people stayed in
  one place.

Small groups of hunter-gatherers often
 have low impact; move on when local
 resources become scarce and
 resources can regenerate.
Some societies practiced sustainable
  agriculture, using selected areas,
  moving on, and allowing these areas to
  go through normal successional stages.

But as agriculture became more
  sophisticated and permanent, ability of
  ecosystems to regenerate diminished.

And -- demand for resources continues to
  grow.
Europe was essentially deforested by the
  1700s, except for land held by the rich;
  many areas of China etc. similar.

So, by the time there was a real interest in
  conservation, much of the natural
  habitat was lost.

North America: Originally aboriginal
  peoples; probably low impact (except
  hunting of large mammals?).
Then: European colonization, exploitation
  of forests and other resources.

And -- a lot of resources were sent back to
  Europe, where the demand was much
  greater; exploitation was no longer just
  local (and this continues to be a
  problem throughout the World).
Early approaches to conservation biology
  in the U.S.:

• Romantic-Transcendental Conservation
  Ethic: Rooted partly in the writings of
  Thoreau, Emerson, Muir.

Philosophical view that nature is the work
  of God -- a “temple” -- not just for
  human/economic benefit. Today, this
  view is exemplified by groups such as
  the Sierra Club.
2) Resource Conservation Ethic:
  Formalized around the beginning of the
  20th Century by Pinchot, following J.S.
  Mill and others.

View of “useful” versus “non-useful”
  components of nature (an
  anthropocentric approach).
Multi-use concept: get as much as
  possible out of an area (timber, fish,
  grazing, etc.); develop management
  strategies to achieve this.
Still seen with groups like the Forestry
  Service, fisheries management
  agencies, Ducks Unlimited etc.

This established the “preservationist”
  versus “utilitarian” schools of thought.


Then 3) Leopold’s Evolutionary-Ecological
  Land Ethic (EELE): mid-20th Century.
Still seen with groups like the Forestry
  Service, fisheries management
  agencies, Ducks Unlimited etc.

This established the “preservationist”
  versus “utilitarian” schools of thought.


Then 3) Leopold’s Evolutionary-Ecological
  Land Ethic (EELE): mid-20th Century.
EELE was rooted in the emerging science
 of evolutionary ecology.

Can’t just break nature into “useful” and
  “non-useful” elements, and need to
  understand how it works.
Study various components of ecosystems
  and their interactions to protect them as
  a whole in an informed way.

Larger an “equilibrium” view (shifted more
  to non-equilibrium as knowledge grew).
The EELE provided the broadest and most
  useful approach, and much of the
  fundamental basis of modern
  conservation biology (although now,
  there’s much more emphasis on the
  human element as well).
1960s-1970s: Biologists started to wake
  up to the fact that entire ecosystems
  were disappearing; biodiversity was
  disappearing rapidly; also, pollution and
  other human-induced problems gained
  wide awareness.

Conservation efforts at the time were
  largely utilitarian or focused on
  appealing species; need for an
  ecosystem approach became evident.
1980: Soulé and Wilcox published
  Conservation Biology: An Evolutionary
  Perspective

Major turning point, followed by other
 works that emphasized evolutionary
 biology as much of the basis for
 conservation.
1985: Society for Conservation Biology
  (and their journal Conservation Biology)
  were established.

Provided a venue for evolutionary and
  genetic approaches to conservation.

The field continues to grow, integrating a
  wide range of areas of science
  (including new genomic tools) and study
  of human activities.
Our text: Three guiding principles that
 establish a paradigm (“world view”) for
 conservation biology.

1) Evolutionary change: A unifying theme
  throughout biology: explains origins and
  patterns of biodiversity; genetic change is
  central to evolution; don’t want to stop
  evolution -- instead, ensure that
  populations and species experience
  natural genetic change, including
  adaptation, in response to natural forces.
2) Dynamic ecology: Ecological systems are
  rarely at equilibrium, and if so don’t stay
  that way long -- no truly stable point.

External forces -- floods, fires, invaders, etc.
  can “perturb” the system or change it
  drastically; ecosystems are usually
  patchy and shift over time.
This view is directly relevant to factors such
  as preserve design; e.g., habitat corridors
  to allow movement and gene flow, and
  consideration of temporal factors.

And of course -- ecosystems are inherently
  dynamic because evolution of the
  organisms within them is happening all
  the time.
3) Human presence:

Can’t leave humans out of the picture; we’re
  here to stay (?), and people won’t support
  conservation efforts if natural areas are
  simply barricaded off etc. Need to
  consider human needs, educate/build
  pride in local habitats, tie conservation to
  economic incentives, recognize rights of
  native groups -- and also, incorporate the
  knowledge that they have of the
  ecosystems and organisms.
Conservation biology is inherently
  interdisciplinary, and also inherently
  inexact in many ways: blending complex
  systems, human components, “hard”
  versus “soft” science, public policy etc.

And, since ecology and evolution are not
  strictly predictable, there’s a strong
  component of probability, plus the need to
  consider many factors that could alter
  expected outcomes -- need to build in
  safety margins.
1.9 The interdisciplinary nature of conservation biology
Chapter 2: Global biodiversity: Patterns
and processes
Biodiversity (biological diversity -- here
diversity is used in a broader sense than
we’ll use later):

Variation across Life.

What kind of variation?
Essay 2.1 (A) Compositional, structural, and functional attributes of biodiversity
1) Genetic diversity: Determines every level
of biodiversity.

Number of genes ranges from a few
(viruses), to several hundred (many
bacteria) to tens of thousands (e.g.,
humans: 20,000 +).

Genetic variation is constantly arising
(mutation, recombination): essential to
evolution.
Levels of genetic variation are often
considered indicators of the “health” of a
population or species: may provide
resilience to changing conditions.

Also critical in captive breeding and
management.
Understanding genetic diversity is
essential to understanding gene flow,
population structure, intra- and
interspecific interactions, relationships to
environment, species boundaries and
evolutionary history...
Population level diversity:

Describes, in part, nature and distribution
of genetic variation within and among
populations.

e.g., local adaptations; species-wide
variation (disease resistance, nutrient use,
etc.

And: genetically-determined phenotypic
plasticity may be very important.
Prioritization of populations for
conservation: maximize diversity?

Different populations may play different
roles in different ecosystems (may or
may not be genetically determined).

e.g., a pollinator, predator, etc. may be
crucial in some systems: not always
obvious until after it’s gone.
3) Human cultural diversity:

Cultures evolve too, and interactions with
environment vary tremendously (good
and bad): many have a long history of
managing/sustaining natural resources.

How many human cultures are there?
Over 6500 known languages (one
indicator).
Can also measure by number of
indigenous populations: generally highest
in tropics (where biodiversity is usually
highest too).

So: complex human/environment
interactions; these determine
cultural adaptation.

Often overlooked, especially in context of
conservation strategies.
2.1 Linguistic diversity and numbers of indigenous cultures across the
world (Part 1)
2.1 Linguistic diversity and numbers of
indigenous cultures across the world (Part 2)
4) Species-level diversity:

For now, just think of number of species in
a given area (more correctly, this is species
richness).

Species are often viewed as the
“fundamental” units of evolution (many
would argue populations). But certainly
they are the “biggest” (most inclusive) units
of evolution.
Higher level taxa (genera, families...phyla,
kingdoms, domains) do NOT evolve.

These are collections of evolving species,
designated by humans.

Boundaries of higher-level taxa are largely
arbitrary, although nearly everyone agrees
that they should reflect evolutionary history
and be monophyletic (= ancestor + all
descendants).
2.2 (A) Domains of biodiversity: Bacteria, Archaea, and
Eukarya
2.2 (B) Major groups of plants
2.2 (C) Major groups of animals
Much of conservation biology is species-
focused: US Endangered Species Act
(ESA); Convention on International Trade
in Endangered Species (CITES); many
others, often more localized.

Danger: Can ignore what’s happening at
the population level; lose genetic diversity
in the wild and in preserves/captive
populations.
What is a species? Many views; can be
extremely important from a conservation,
political, and legal perspective.

Very important to distinguish between
CONCEPT and PRACTICAL
APPLICATION.

Species recognition/designation is a
hypothesis, and often testable.
“Classic” concept: Biological Species
Concept (Dobzhansky, Mayr):

“A species is a group of actually or
potentially interbreeding populations
which are reproductively isolated from
other such groups”.
BUT: Problematic in many ways.

Practical: Often very hard to test, especially
with non-sympatric organisms.

Fundamental: Lots of species interbreed
to varying extents (and reproductive
compatibility could just be a retained
ancestral condition).

What about asexual lineages?
Very restrictive: if two groups are sympatric
(live in same area) and don’t ever
interbreed -- yes, they are different species.

Or if they interbreed and offspring are
completely inviable or sterile -- different
species.

But what about the in-betweens? Say,
slightly reduced fitness (or even higher
fitness of hybrids).
Over 99.99% of all species on Earth are
recognized based on indirect evidence of
reproductive isolation: morphology,
behavior, etc. (or simply no consideration
of reproductive isolation).
Phylogenetic Species Concept:

Not really “phylogenetic”; simply requires
that a group be monophyletic,
interbreeding, and display some
feature unique to the group (an
autapomorphy).

But -- in practice, almost any population
could be a species. And, do species
have to be distinguishable at all?
(Conceptually)
Evolutionary Species Concept (Simpson,
Wiley and others):

This is a real concept. A species is
“a single lineage of ancestral-descendant
populations of organisms which maintains
its identity from other such lineages and
which has its own evolutionary
tendencies and historical fate”.

(A lineage concept).
What are the implications?

Maybe we’ll never be able to detect all
species -- sometimes we have sufficient
evidence, sometimes not.

Unsatisfying to many (including lawyers!)
who want some simple, quantifiable
measure (say, genetic distance).

But there isn’t one.
Are species real entities?

Some would argue that they’re just
human constructs, it’s all a continuum,
etc. But obviously there are distinct
lineages with distinct identities.

If you don’t think that species are real,
then you can make up various criteria --
but they will never hold up across
different groups.
In legal situations, it’s common to invoke
the Biological Species Concept -- which
is rarely testable (so supposedly, if its
requirements can’t be satisfied, then a
given unit isn’t “proven” to be a distinct
species).
And -- back to the issue of what to
protect -- if the goal is genetic diversity,
evolutionary potential, etc., then
populations etc. may even be more
important.

The ESA does allow for protection of
distinct “subspecies” and populations
(Evolutionarily Significant Units, ESUs),
but just for vertebrates, and generally
less valued than full species.
And, if populations are the units of
evolution, if there’s local adaptation,
geographic genetic variation, if different
populations play different roles in
different ecosystems --

Then, protecting just one or a few select
populations (or captive breeding) may
provide a false sense of security.
Species-only focus can lead to loss of
genetic diversity: consider crop plants:
may have lost resistance to disease (or
that ancestral population didn’t have it):

If the wild ones are gone, usually
irreversible.
Pros of species emphasis:

Allows focus on particular entities, and a
basis for estimating biodiversity.

Can appeal to public (e.g., Giant Panda
etc.) and at the same time protect whole
ecosystems.
But --

Problematic if the approach obscures
attention to ecosystems and results in
only select species being protected in
isolation of their ecological context.

Captive breeding is an especially strong
example.
Have to be careful, too, that species
aren’t overlooked.

Especially with molecular genetic
advances (DNA sequencing etc.) many
new species are being discovered that
were “cryptic” (e.g., not identifiable
based on morphology).

So one “species” often is two, or even
many (salamanders are a great
example).
How many species on Earth?

Described (= formally recognized): about
1.75 million living, plus about 300,000
fossils.

Dozens to hundreds described every
day; hundreds to thousands probably go
extinct every day, before we even know
that they exist.
Wide range of estimates: about 10
million - 100 million (text says high end is
50 million). Probably more.

Fungi: About 80,000 described species;
1 million+ suspected.

And viruses, bacteria, archaeans, etc. --
just beginning to scratch the surface
(even though these may be critical to
ecosystems worldwide).
e.g., fogging of tropical forest canopies
reveals huge number of species
previously unknown (55% new); think of
the oceans!

Even vertebrates: salamanders in central
Texas: at least 12 species when there
were only 4 recognized until recently; 4
new ones described in last few years,
some right in the city of Austin and
nearby urbanizing areas.
Table 2.1 in text:

Bacteria: About 5,000 described species;
around 1 million expected.

Nematodes: About 25,000 described;
around 400,000 expected.

Arthropods: About 1 million+ described;
around 9 million expected.

Chordates: 58,000; 60,000
Diversity of higher-level taxa:
classifications are controversial, but at
least 3 Domains: Bacteria (= Eubacteria),
Archaea, and Eukaryota; many biologists
recognize 20+ Kingdoms (previously 5 at
most)

Preservation of higher-level taxa: even
though these are human constructs,
usually distinguished by “key
innovations”; lose the whole group and
lose a key part of biodiversity
2.2 (A) Domains of biodiversity: Bacteria, Archaea, and
Eukarya
Consider genetic diversity using
evolutionarily ancient small subunit
ribosomal RNA gene: sequence diversity
seen in animals, true plants, and fungi
only about 10% of that seen across the
rest of Life.

Origins of many groups most familiar to
us are relatively recent.
Deeper splits (e.g., among Domains) are
billions of years old (Life arose at least
3.6 billion years ago, not long after Earth
became habitable).
Biological communities: often hard to
define:

Species + interactions with each other
and the environment.

Changes across landscape with abiotic
factors etc.
Species richness and species diversity:

Richness = simple count of number of
species (often a selected group).
Includes both rare and abundant
species.

Diversity: Weighted measure of species
“evenness”, based on numbers,
biomass, productivity, etc.
e.g., if all species equally abundant, high
degree of evenness.

Various indices of similarity in diversity
across ecosystems; provides an
indication of habitat types (e.g., high
proportion of one tree species may be
predictive of the kind of ecosystem,
versus similar abundances of many
species).
Richness can be measured on different
scales: “alpha richness” describes
number of species in a given area.

“Beta richness” describes change in
species composition (cumulative
number, or turnover -- i.e. change in
species composition along a gradient).

“Gamma richness”: Larger scale
extension of beta richness.
Alteration of habitat, climate change,
pollution, and other human-caused
(anthropogenic) changes often lead to
changes in both species richness and
diversity.

“Weedy”, resilient species take over:
raccoons, blue jays, tough plants; often
introduced species: less richness, more
evenness (some species may persist,
but their numbers become low).
BUT: Many biologists still focus on
richness, partly because rare species may
be very important (“keystone species” and
“ecological engineers” for e.g.).

These may be more vulnerable to being
completely lost, and are often harder to
census.

Note: indices of richness and diversity
rarely distinguish between native and
introduced species.
Importance of particular species is often
hard to assess, and understanding how
a community or ecosystem works is
extremely complex and difficult.

This is a key reason why conservation
biologists often focus on selected
species (and often hope that this will
preserve the whole ecosystem).
Richness/diversity of biomes and
ecosystems worldwide: traditionally
focused on plant composition (and later
climatic factors too).

Eight major terrestrial biogeographic
regions are generally recognized:
1) Palearctic (northern part of Old World,
   from Europe and N. Africa through N.
   and N.-central Asia).

2) Afrotropic

3) Indo-Malay

4) Australasian
5) Oceania (S. Pacific islands)

6) Nearctic (Arctic through Caribbean
and Mexico)

7) Neotropical (New World tropics)

8) Antarctic
Each has ecological gradients: consider
temperate latitudes: can follow
precipitation gradients.

Wet forest to woodland to desert; boreal
forest in north,then tundra: call these
ecosystem types biomes.

Tropics: Rainforest, evergreen seasonal
forest, dry forest, thorn woodland, desert
scrub, desert.
2.4 Biomes and climate
Elevational gradients: e.g., lowland
rainforest, montane rainforest, cloud
forest, elfin woodland, paramo (alpine
tundra above treeline).

Biomes can be further subdivided by
slope, soil type, species present, etc.

Of course, hard to define precisely.
2001: World Wildlife Fund recognized
867 terrestrial “ecoregions”: large areas
with key communities, ecological
characteristics, and usually some major
groups of species spread through the
region.

More than 1000 researchers involved;
the most comprehensive basis for
classifying landscape-level
richness/diversity on a broad scale.
Each ecoregion has a variety of habitat
types; try to preserve as much of each in
each ecoregion to maximize richness
and biodiversity.

Major attempts are underway to to
classify marine ecoregions: their
ecosystems are extremely important but
still little understood.
Geological history of species richness:

Determined by rate at which species
arise vs. go extinct.

Various factors may promote speciation;
vicariance (geographic splitting) is
arguably the most important.

And -- the more species there are,
And -- the more species there are, the
more (theoretically) chances there are
for further speciation: feedback and
exponential effects.
Loss (extinction) rates driven largely by
disturbance (although some level of
disturbance may promote speciation and
species richness).

Also competition, predation, disease, etc.
Life arose at least 3.6-3.8 BYA:
prokaryotic, then eukaryotes too by
about 2 BYA.

Species richness probably low until
around 1.6 BYA (late Precambrian -- but
early organisms probably didn’t fossilize
well).

Mass extinctions: try to distinguish from
“background” extinction rate.
Then, arthropod and echinoderm-like
animals appear in the fossil record, plus
other lineages that probably have gone
extinct.

Early Cambrian (approx. 550-600 MYA):
“explosion” of speciation? Also, more
organisms that fossilize better (e.g.,
shells).

Another pulse starting about 450 MYA.
Five “major” extinction events often
recognized (plus the sixth happening
now).

A big one: Permian extinction approx.
250 MYA: Pangaea breaking up,
currents, climate changing, possibly
asteroid impact: estimated 95% of
species lost (note that next figure
estimates loss of families).
Another big one: Cretaceous extinction
at K/T (Cretaceous/Tertiary) boundary
approx. 65 MYA.

Asteroid impact; estimated 75% of
species extinct, including most
dinosaurs.

On average, species richness seems to
recover in 10+ MY. And, massive losses
may open up opportunities for new
species; overall, richness has increased.
Terrestrial vascular plants by about 400
MYA; major explosion of angiosperms
starting approx. 150 MYA, then
especially 65 MYA.

Appears to be correlated with explosion
in insects: maybe each promoted the
other.

Diversification of angiosperms may have
fueled that of terrestrial animals (and
other organisms).
Endemism: Localization of a species to a
particular area (some bacteria are
endemic to most of the Earth; other
species may be endemic to just a tiny
island).

Largely the result of vicariance
(geographic separation); also
specialization for particular
environmental conditions; a particular
group may be endemic to an area if the
ancestor gave rise to multiple species.
Islands have high levels of endemism,
but not necessarily richness (depends
partly on size of island).

But: oceanic islands with coral reefs
have high richness and endemism -- for
example, western Pacific islands
Patterns of endemism greatly among
taxa in a given area; species with high
dispersal may be widespread, and
intrinsic features of the organisms may
promote localization, due to resource
requirements or the nature of speciation
Cape of South Africa, southwestern
Australia: low endemism of mammals
and birds, extremely high endemism of
plants.

May be due to polyploidization: fairly rare
in animals, common in angiosperms:
promotes localized speciation.

Marine systems: much less understood
than terrestrial.
2.10 The Indo-West Pacific is a
marine diversity hotspot
Latitudinal gradients in richness:

For most groups, species richness
highest closest to the equator -- true for
terrestrial and marine organisms.

Ants: Locally, around 10 species far
north/south, up to 2000 in some areas of
tropics; breeding birds: Greenland 56,
NY 105, Colombia 1395; bivalves (see
next figure).
Number of species, genera, and families of
              marine molluscs by latitude
Reason for latitudinal gradient still
unclear; probably combination of factors:

-speciation vs. extinction rates?

-habitat heterogeneity?

-energy availability (productivity)?

-levels of gene flow?
In general, a major determinant of
species richness is area: highly
correlated with speciation and extinction
rates.

First noted by Arrhenius; true of both
islands and whole continents.
Species - area curves:

S = cAz

Where S = number of species, A = area,
c and z: constants fitted to area

On log scale, linear: c = y-intercept and z
is slope

Next figures: Caribbean amphibians and
reptiles; Pacific birds
Generally, z is about

-0.15 for continents

-biotic provinces within a continent: 0.19

-islands associated with continents range
from about 0.25-0.45 (varies partly
according to proximity to mainland =
colonization, or islands that periodically
connect to mainland)
MacArthur and Wilson (1960s): Theory of
Island Biogeography: predict number of
species based on island size, proximity
to mainland, colonization vs. extinction
rates.
Consider areas of climatic similarity:
largest in tropics:

-little change in temperature at middle
latitudes

-middle latitude bands have higher
surface area

- larger area: higher speciation rates,
lower extinction = high richness
Perhaps more stability in tropics; also
generation time effects/frequency of
reproduction.

More generations = more genetic
variation = faster evolution = more
speciation?
Energy effects: the more energy from the
sun, the more biomass per unit area can
be generated; higher packing of
individuals (and probably species).

High energy input (especially with high
moisture) may put less strain on
individuals, allow more of their energy for
reproduction; may even allow greater
dietary specialization, contributing to
speciation.
Feedback effects: more species, more
speciation.

Primary productivity: primary source of
energy available to organisms; a function
of temperature, moisture, nutrients.
2.14 Numbers of species of Eucalyptus and Acacia in southwestern
Australia
2.15 Plant species richness in Borneo (1-3) and in
              southwestern Australia (4-5)
Number of species of marine molluscs, by salinity
Effects of disturbance on species richness:

Major, cyclical disturbance: climate change
(except that now it’s probably permanent):
especially strong in N and S latitudes.

If it happens fast, species can’t adapt; life
history disrupted, ranges shift, extinction
rate increases.
Promotes wide individual ranges,
generalization (food, temperature, etc.);
increases gene flow, decreases
speciation rate (while extinction rate is
increasing).
Magnitude and and localization of
disturbance is important:

Small: storms, floods, disease,
competition, predation etc.: generally
localized; if extreme, richness may suffer
but can probably recover.

Widespread and major: maybe not, at
least in the short term.
Some argue that stability of habitat
allows species to accumulate (e.g., deep
ocean floors): productivity low, but
patches of resource-richness (e.g.,
hydrothermal vents).

So -- promotes genetic isolation,
speciation; little disturbance to cause
extinction.

Maybe the same for the tropics (perhaps
a long history of stability?).
Interactions among organisms:
competition may reduce alpha richness if
one or a few species take over.

BUT: Can also promote richness if it
encourages niche partitioning.

Depends on resources and population
densities.
Predation: Classic example: sea stars on
Pacific coast.

Eat mussels that otherwise would
dominate the community.

Sea stars present in low numbers, but
keep competition in check (remove sea
stars and richness decreases.

Similar: grazing of prairie plants by
buffalo may promote richness.
Intermediate Disturbance Hypothesis:

Disturbance by changes in abiotic or
biotic environment (especially from
outside sources), at medium levels,
prevents takeover by a few species.

Also increases diversity in the sense of
promoting evenness.
More extreme disturbance: Knocks back
richness (eliminates some species,
directly or indirectly causes increased
competition etc.).
And, frequent disturbance may prevent
individuals from maximizing reproductive
potential.

So intensity and frequency of
disturbance contribute to richness (and
diversity) (“optimum” varies by
ecosystem).
Intermediate disturbance hypothesis
Local and regional species richness:

Ecological communities are shaped by
local species interactions and
environment, but broader-scale
processes are critical on time and spatial
scales.

Local: mutualism, competition, predation;
large scale: biogeographic history,
phylogeny, dispersal, speciation...
Is there a limit to species richness?
Maybe, for a given community:
interactions plus abiotic environment
may be constraints.

May be a limit to how specialized a given
species can be re: resources; limit to
levels of niche overlap.

Many other factors can be limiting:
pollination, parasitism, predation, etc.
Many groups show pattern of increasing
alpha (local) richness with increasing
regional (beta) richness, but rate of local
richness increases more slowly.

For many systems, local interspecific
interactions may be most important in
determining alpha richness, and relative
abundances of species (and thus,
measures of diversity).
Importance of “biodiversity”: value to
humans is a practical consideration.

e.g., number of species of wild bees
directly determines crop pollination
efficiency: preserve nearby habitat;
nutrient cycling key to human activities;
medical compounds, etc.
Chapter 3: Threats to Biodiversity



Humans are the biggest direct and
indirect threat to biodiversity (besides
occasional asteroid impacts etc.).
3.1 Major forces that
threaten biological
diversity
Major categories of anthropogenic
 effects:


1) Habitat degradation: Pollution, change
   of native habitat, introduced species,
   fragmentation, complete destruction.
3.6 Habitat loss and degradation is the greatest threat to global
biodiversity among these groups
Major categories of anthropogenic
 effects cont’d:


2) Overexploitation: Logging, fishing,
  medicines, pet trade: may eliminate
  species or reduce numbers to the
  point of genetic inviability.
3) Introduced species: Sometimes a
  component of 1).

e.g., introduced plant takes over and
  changes habitat.

Also competition, predation (e.g., brown
  tree sanke in Guam has wiped out
  some native birds); disease; changes
  in interactions among native species.
4) Human-induced climate change:
  Previous mass extinctions often tied to
  major climate change -- continental
  drift, impact of extraterrestrial objects.

Climate is now changing fast, and
  change is accelerating: population
  growth, industrialization.
Feedback effects: e.g., Arctic ice melts,
  exposes soil, absorbs more heat, more
  decomposition, releases CO2...

Cumulative/synergistic effects of
 changes: one pollutant may not be a
 big problem, but can increase
 vulnerability to disease, parasites,
 other pollutants.

And: bioaccumulation of toxins in food
  chain/web.
3.2 Toxic chemicals that
accumulate in fatty tissues
concentrate at the top of the
food chain
Essay 3.1 (A) Transient killer whales represent
the most PCB-contaminated marine mammal
(white = male, gray = female). Transients eat
marine mammals; residents eat fish; females
pass PCBs to offspring via milk.
Corals: symbiotic photosynthetic “algae”:
warm water and/or pollutants cause
  symbiont to leave.

Coral bleaching; corals die, and so does
  the whole reef ecosystem.
Cascade (snowball) effects: Consider
  invasion of the oomycete
  Phytophthora in Australia: root parasite
  that kills weedy species.

Herbaceous plants take over and
  smother seeds of woody plants; this
  changes the community of animals
  and other organisms too (e.g., grazers
  and plant-eating insects increase, but
  others decrease; nesting reduced,
  etc.).
3.3 “Snowballing” effect of invasion of an alien root
pathogen (Part 1)
3.3 “Snowballing” effect of invasion of an alien root
pathogen (Part 2)
Combined effects: Many island biotas
  decimated by both human effects via
  hunting and habitat destruction, and
  also introduced species (rats, pigs,
  goats...) that came with them.

Hawaii: People killed off many native
  birds (largely for their feathers); many
  were ground nesters vulnerable to
  human-introduced species; few native
  predators before that. Plus
  anthropogenic habitat change.
Almost 90% of large mammal genera
  were lost in Australia after human
  colonization approx. 50,000 + years
  ago. Similar in North America, Middle
  and South America (10-15,000 years
  ago).

May have been tied to climate change as
 well: warming 14-10,000 years ago,
 then glaciation in the North.
3.4 Mammalian megafaunal genera
went extinct soon after human migrations
Extinction can be local (extirpation),
  global (lose entire species), or
  ecological (numbers reduced, can’t
  play key role in ecosystem; e.g.,
  pollinator, top predator).
Extinction can also be “genetic”, or
  caused by change in genetic
  composition, especially loss in
  variation.

Reduce population size and gene flow:
  genetic drift, inbreeding: “extinction
  vortex”.
3.5 Keystone and dominant species can have large
impacts on biological communities
Sea otters: Prized for fur in 19th and
  early 20th centuries.

Eat sea urchins in Pacific kelp forests.

Sea urchin population exploded, and
  huge amounts of habitat for many
  species were lost.
Lose predators such as mountain lions
  and deer populations explode.

Overgrazing, effects on habitat, less
 obvious effects.

e.g., in central Texas, deer feed on bark
  tree that Golden-Cheeked Warblers
  use to make nests when other food is
  depleted.
Some species may not be abundant, but
  still very important:

”Keystone species”: effect on community
  disproportionate to numbers or
  biomass (e.g., top predators,
  pollinators, seed dispersers).

”Ecosystem engineers”: beavers,
  elephants.
Others dominate and shape the
  ecosystem: trees, corals, plants that
  prevent erosion, etc..

Also decomposers (bacteria, archaeans,
  fungi, and others): don’t often think of
  these.
Estimated numbers of globally
  threatened species: tracked by IUCN
  (International Union for the
  Conservation of Nature and Natural
  Resources): Red List.

Levels of risk: Least Concern, Near
  Threatened, Endangered (there are
  subsets of these too).
Box 3.2 (A) IUCN Red List categories
41% considered Threatened to some
  extent; over 800 considered Extinct; a
  few only exist in captivity.

Obviously a huge underestimate:
 focuses on known and conspicuous
 species, and how many do we have
 historical (or even much current)
 information on?
Threats:

Biggest: habitat destruction, degradation,
  fragmentation (and can include
  pollution here). Includes migratory
  species: e.g., some birds, monarch
  butterflies (overwintering grounds; next
  figure is monarch butterflies).

For some, overexploitation: e.g.,
  fisheries.
Fall and spring migrations
of the monarch butterfly
Forest cover has dropped dramatically within and adjacent to the
Worldwide, see the largest numbers of
 threatened species in areas of
 greatest species richness: moist
 tropical forest, tropical grassland; also
 montane regions (which may be
 especially affected by climate change).
Proportion of threatened species in North
  America (in the broad sense) is high,
  partly because of southern subtropical
  and tropical regions (and especially
  Hawaii).

Plus, huge amount of agriculture,
  industry, urbanization, with a long
  history.
Plus, we know a lot more about North
  American species than those in SE
  Asia, Amazon, Africa, etc. -- there,
  many species are being lost before we
  even know about them.
3.7 Species richness of mammal, bird, and amphibian species in the major biomes of the
world (black = threatened)
Threats in the U.S. by taxonomic group
  and category (next table):

About 1200+ species are listed as
  Threatened or Endangered; some
  analyses suggest that up to 1/3 of U.S.
  species risk extinction.
U.S. started to address this relatively
  early:

1966: Endangered Species Preservation
  Act (native species).

1969: Endangered Species Conservation
  Act: extends protection to worldwide
  species (regulates imports etc.).

1973: Endangered Species Act (ESA).
ESA is more powerful than previous
 ones: extended protection to plants
 and invertebrates.

Idea of “critical habitat” and more power
  to acquire land.

Requires that all federal agencies follow
  the rules.

Better integration with state conservation
  authorities, plus funding.
Various amendments: a key one is that
  economic concerns cannot take
  precedence over protection.

Use biological, NOT economic
  information (rule isn’t always followed).

Snail Darter in TN in 1970s: threatened
  by dam building: eventually exempted
  from ESA even though Supreme Court
  favored the fish.
Under ESA:

Threatened: In danger of becoming
  Endangered.

Endangered (imminent danger of
  extinction, with subcategories).

Plus: Species of Concern: being watched
  closely.
Taxonomic patterns in the U.S.: lots of
  plants (but of course, there are lots of
  known species).

However, freshwater species such as
  many invertebrates, fishes, and
  amphibians top the list, with (percent-
  wise) freshwater mussels worst off.
  Water resources are at great risk.
3.8 Proportion of species threatened with
extinction by plant and animal groups in the
U.S.
Elsewhere in the World:

Number of species under threat largely
  correlated with amount of research:
  e.g., intensive study of plants in
  Ecuador, and many turn out to be at
  great risk.

Similarly, Madagascar is a big focus:
  hugely deforested; more than 2/3 of
  plants and 1/3 of vertebrates are in
  serious danger.
China: Major problems:

Habitat loss is a huge issue, but
  overexploitation is especially bad
  (food, medicine).

Turtles: thought to prevent aging, cure
  cancer, etc.

Some markets have literally tons for
  sale, and some are extremely valuable
  ($1,000s).
So many lost that now major imports
  from Vietnam, Indonesia, India, U.S.,
  even Madagascar.

Several species are known only from the
  markets; probably have (or had) very
  limited ranges; some don’t appear in
  markets any more.
Little chance of recovery: turtles are
   slow-growing, long-lived, usually age
   to maturity is long, not many offspring,
   often have specialized habitats and
   requirements.

This also limits “farming” of many
  species (although becoming more
  economical).

Similar issues with mammals, birds, non-
  avian reptiles.
3.9 Contribution of major threats
to endangerment among vertebrate
species in China
Recall that species richness increases
  with area: S = cAz, where S = number
  of species, A = area, and c and z are
  constants fitted to the data.

On a log10 scale, the relationship is
 positively linear with z (slope) ranging
 from about 0.15-0.35.

So -- can predict the number of species
  lost with decreasing area.
Box 3.3 (A) Species–area relationships based on
the equation S=cAz, with c=10
So, here, reducing an area of 1000, say,
  hectares to 100 should reduce the
  number of species by approx. 38%
  (z=0.20) to approx. 55% (z=0.35).

Allows some prediction of effects of
   habitat loss, taking into consideration
   the kind of environment (considered as
   an “island).
Tropical forests: losing about 0.5% per
  year; assume that z = 0.2: then, about
  0.125% of species lost per year.

Suppose 5 million species; lose 6,000+
  species/year.
Estimate (very rough) of the
  “background” rate of extinction is 10
  species worldwide/year (tons of
  assumptions, and doesn’t include
  mass extinctions).

Estimated current rate of species loss is
  much higher (50-10,000X).
Recovery rate (if factors causing
  extinctions stopped) can be very high.

Fossil record and observations suggest
  that to regain the species richness of a
  lost coral reef could take up to 10
  million years.
Vulnerability of species:

Some have characteristics that make
  them more vulnerable than others.

-need large area of contiguous /
  connected habitat (e.g., large
  mammals)

-need limited type of habitat
- restricted to small geographic area

- food specialization

- codependence on other species,
  coevolution

- low reproductive rate

- rarity (discuss in a minute)
r vs. K selection:

r-selected: often in unstable
   environments; reproduce at early age,
   lots of offspring, don’t invest much in
   them; most don’t make it but a few do.

K-selected: longer-lived, slower
  maturation, fewer offspring but more
  investment in each, may not reproduce
  unless conditions are just right.
And, since generation times for K-
  selected species are longer, takes
  longer to adapt to changing conditions
  (plus not as wide a range of genetic
  variation in offspring).

So: Habitat degradation, exploitation,
  etc. may lead to a shift to species that
  are r-selected; can change whole
  structure of ecosystem (and perhaps
  reduce richness).
What is “rarity”? Can mean lots of things.

Widespread but not abundant in any one
 place.

Abundant but very restricted distribution
  and/or habitat.

Most vulnerable: Limited range, habitat
 specialization, small population sizes:
 threatened by catastrophe,
 exploitation, genetic deterioration.
Centinela Ridge (Ecuadorian Andes):
  RAP (Rapid Assessment Program)
  inventory 1980s: high levels of species
  turnover (high beta richness), many
  endemic species.

Had at least 90 species of endemic
  plants; just after study finished,
  converted to agricultural land; all
  endemic plants went extinct.

Common situation, especially in tropics.
“Artificial rarity”:

Suppose that a species is widespread
  and abundant; then experiences
  reduced range and population size.

May not be able to adapt to this (e.g.,
 birds such as Passenger Pigeon that
 nested in huge numbers -- part of their
 life history). Or, loss of genetic
 variation.
In some cases, a species with specific
   habitat requirements and/or low
   population size may actually be better
   suited to some further reduction.

And, if localized but widespread, may
  escape loss of at least some
  populations.
Percentage of threatened U.S. species by taxonomic group
3.11 The Living Planet Index (LPI) tracks population trends
for over 1100 vertebrates (Part 1)
3.11 The Living Planet Index (LPI) tracks population trends
for over 1100 vertebrates (Part 2)
Prediction of threats to species and
  ecosystems:

Biggest: Habitat loss, climate change,
  pollution.

Overexploitation and invasive species
 lesser threats, but will increase, just
 not as fast.
The ONLY trend likely to slow is
  destruction of temperate forests.

But this is just because most are already
 gone, and much of such habitat is in
 areas where people are affluent
 enough to preserve what’s left (or
 there just aren’t that many
 people/access difficult).

Need education & creation of incentives.
3.12 Projected trends in threatening processes in different
habitat types (Part 1)
3.12 Projected trends in threatening processes in different
habitat types (Part 2)
Amphibian declines: Well studied; proxy
 for other groups too.

Late 1980s: Biologists started to notice
  that various species of frogs and
  salamanders weren’t nearly as
  common as they used to be -- some
  even seemed to have disappeared.

e.g., Golden Toad in Costa Rica; Gastric
  Brooding Frog in Australia.
And: Many of these declines were in
 seemingly pristine, relatively
 undisturbed areas (especially
 montane).

Species-specific, but worldwide: Global
  Amphibian Assessment (mid-2000s
  and ongoing) found that about 1/2 of
  400+ species that were declining fast
  were doing so for unknown reasons.
Amphibian threat status worldwide
Most are tropical; South America and
 Australia best studied; particular
 groups of frogs seem especially
 vulnerable.

Possibilities:

1) UV radiation: Bad for eggs of species
   that lay in areas of sunlight.

May act together with parasites, acid
 rain: synergistic effects.
Depletion of ozone layer allows more
  UV-B through.

Global warming makes breeding ponds
  shallower; less rain in some
  areas/more evaporation.

Add acidification (and maybe toxins):
  features such as plants, microbes etc.
  change.

Different plant composition: less shade.
Less water: amphibians congregate
  around water sources: greater spread
  of disease.

2) Disease: 1990s: chytrid “fungus”
  discovered that attacks amphibian skin
  -- seemed to correlate with declines in
  Australia and Central America.

Slows growth rate of tadpoles/weakens
  or kills adults (maybe salamanders
  too).
Developed rapid PCR (polymerase chain
  reaction) test for chytrid-specific DNA:
  seems to be associated with some
  declines in N. America, Europe, other
  parts of the world.

BUT: Present in some species without
 causing harm: natural resistance?
 Environmental interactions?

Rare in tropical lowlands: does best in
  cooler temperatures.
Could be ancient, with cycles of
  epidemics?

Maybe environmental stress makes
 some amphibians more susceptible?

OR -- Could be new, and recently
 introduced to some amphibian
 populations (introduced species,
 humans, other carriers).
Disease cycles (in general) are normal,
  but why simultaneous declines
  worldwide and number of species
  affected in different habitats.

PCR testing doesn’t show it in older
  museum collections: sudden
  appearance.
Also -- very little chytrid DNA sequence
  variation -- suggests very little
  isolation, past or present; recent
  spread.
Maybe carried by introduced N.
 American bullfrogs and especially
 African Clawed Frogs (Xenopus) --
 widely introduced around the world.

And -- first documented in Xenopus,
  which appears to tolerate it well
  (perhaps coevolved mutualism or
  commensalism).
Other causes of declines:

3) Introduced species; e.g., trout
  stocking; accidental introduction of
  other species of fishes.

Competition with/predation by other
  frogs: bullfrogs, marine toads, Cuban
  treefrog.
California: Introduced crayfish (fish bait)
  destroy eggs and larvae of
  salamanders that have natural skin
  toxins; crayfish are immune (crayfish
  are bad news in many areas).

4) Pollution: Some chemicals disperse
  even to remote areas: DDT, PCBs.

Locally, atrazine (herbicide) causes
  sterility and developmental problems.
Pollution/acid rain may increase stress,
  susceptibility to other factors.

Pollution and climate change can affect
  habitat: e.g., lose trees, less shading
  of streams, ponds (UV effects);
  disruption of fire cycles increases
  waterside vegetation, interferes with
  breeding migrations and metamorphic
  emergence onto land.
Amphibians likely are an indicator for
 more widespread issues, and illustrate
 interconnectedness of threats, as well
 as cumulative/synergistic processes.

But -- sometimes hard to distinguish from
  natural population fluctuations -- gets
  tricky.
Case 3.1 (C) Fluctuations in population size of two species of
salamanders in a temporary pond




      Grey = female; black = larval; 1=l
Madagascar: Huge biodiversity hotspot:
 many endemic species and groups
 (99% of amphibians, 93% of
 mammals).

Separated from African mainland for
  approx. 160 million years; humans
  arrived about 2,000 years ago.

Over 80% of native forest gone; extreme
 poverty, slash & burn agriculture,
 charcoal production.
Case 3.2 (B) Loss of
forest cover on
Madagascar
Native plants very sensitive to
  agriculture, grazing; soils low in
  nutrients; reduce forest cover and soil
  blows away.

Intensive human predation,
   overcollection, many introduced
   species.

Until recently, only 3% of land area
  protected.
Huge interest from international
  community:

Intensive mapping, identification of
   “hotspots” of species richness,
   creation of national parks that
   maximize range of habitat types (or
   preserve unique ones).

Work with local people to create
 incentives for conservation.
Cost/benefit analysis:

Different scales: Geographic, political,
  economic.

e.g., although timber harvesting creates
  many $$$ in the short term, most
  companies are foreign: most profits
  leave the country; forest is gone;
  industry leaves -- government gets
  some short-term benefit (a little bit to
  the average person) but short-sighted.
In the long term sustainable harvest is
better, even just economically (within the
country).

Global impications: Forest removes lots
of carbon from the air: world value about
$500 million/year.

Split incentive: How to convince
Madagascan government to preserve
forests and create/maintain parks?
International governments have given $
millions, but payback has been huge:
protection of drinking water, erosion,
establishment of ecotourism, etc.

Improvements (as of mid-2000s): rate of
forest loss declined to about 1/20th that
in mid-1990s; burning has dropped to
about 1/16th; some ecosystems
recovering (e.g., coral reefs affected by
runoff).
Slash & burn agriculture and charcoal
production still a problem: agricultural
land only has enough nutrients for 2-3
years, then move on; also, still too much
logging.

But: Help from outside to establish
sustainable logging, ecotourism; improve
relations between government and local
peoples (put locally generated $$$ into
local economies; education + technology.
Social sustainability:

Establish systems to oversee the
process and ensure continued benefit of
local peoples: as people see the
benefits, the idea spreads.

Not perfect, but a model for conservation
elsewhere in the world.
Chapter 4: Conservation Values and
Ethics

We’ll only cover this briefly, and
introduce some key points that will come
up again.
Distinguish between instrumental
(utilitarian) value -- what good can
biodiversity do for humans?

And:

Intrinsic value: Inherently valuable,
whether or not useful to humans
(biocentric/egocentric view).
Norton’s (1991) convergence hypothesis:

Idea that anthropocentric (human-
centered) “instrumental values” intersect
with non-anthropocentric “intrinsic
values” (and pure intrinsic value ideas
can actually be damaging to the case for
conservation).

Anthropocentrism covers a lot of ground,
from goods and services to spiritual
meaning.
4.1 Norton’s convergence hypothesis
Taylor’s (1986) biocentrism: All
organisms are inherently valuable, even
individuals. But how to apply? And how
does this relate to ecosystems etc.?

Rolston (1994) added a more complex
“moral” dimension: aggregates
(ecosystems, evolution itself, etc.) are
worth more than species, for example.

Sentient beings are worth more than
non-sentient ones.
4.6 Taylor’s and Rolston’s biocentrisms
4.2 Burden of proof according to instrumental and intrinsic value systems
4.3 Burdens of proof according to the standard CBA and the SMS
approaches
Chapter 5: Ecological Economics and
Nature Conservation


Obviously, people will always need/want
resources; a tradeoff; often requires
sociological, anthropological, political,
and economic perspectives.
No simple answers -- but what if you can
put a price on conservation?

Madagascar: Can place a monetary
value on reduction of atmospheric CO2.

Economists try to assess the effects of
one set of actions versus another on
human interests.

Usually money is the measure.
Concept of utility: “Wellbeing”: How
much does the average individual
benefit?

And how do they benefit? Higher
income, happiness from enjoying
biodiversity, quality of life/health (e.g.,
reduced pollution).

One way to assess utility: how much will
people pay for something?
Text example: Buy a drink; how much will
you spend?

A lot: Net loss in utility (unless it really
meant a lot e.g., extreme thirst); get a
good deal: gain in utility.

Often tradeoffs.
Road construction project:

Try to increase or at least balance utility;
people may be willing to accept
something that improves their utility in
exchange for less or no construction
(e.g., drive further, but less
pollution/noise).

Some of this is unmeasurable (at least
directly).
But some is measurable:

e.g., park land, hunting, fishing, clean
water, health costs of pollution, clean
water, house prices -- things that can
affect the local economy.

Is the utility higher or lower once the
project is completed? If lower, is there
compensation for lost utility (more jobs,
less transit time, some other recreational
area set aside...)?
Cost-benefit analysis (CBA):

Many aspects of utility are intangible,
even philosophical: how to quantify?

“Ecological economics”.

1981: U.S. Presidential Executive Order
requires CBA for new policies/projects.

What’s best for society as a whole?
Of course, not everyone will benefit.

But, a CBA has the “hypothetical consent
of the citizenry”: elected officials should
implement policies that public perceives
improve social welfare.

CBA has 5 steps.
1) Project definition: What is the project
   and how will its utility be assessed?

Suppose an oil company wants to build
  an oil rig.

Goal: Increase profits. Costs to
 company: construction, oil yield, oil
 prices. Environmental concerns
 probably less (although do have to
 consider PR costs).
The government “should” look at the
  other side of things: tradeoffs to public
  welfare. And, what are the
  “boundaries of inquiry” (just like the oil
  company does)? = how many factors
  to consider?

Of course, government doesn’t always
  have opposite interests, and decisions
  have to be structured at different
  levels, e.g., local vs. national.
2) Impact classification: For the
  company, the costs etc.; for the public,
  employment, local economy, also
  environmental costs for life of oil rig.

Consider immediate and long-term
  costs/benefits.

What impact will the project have on the
 environment?
Have to consider what might happen
 versus what’s likely to happen anyway.

Consider example of a rare species of
 finch whose breeding habitat will be
 impacted by construction.

30% of breeding habitat will be lost:
  raises chance of extinction in 100
  years from 10% to 30%.
BUT: Oil company looks at migration:
 70% of migratory habitat estimated to
 be gone in 10 years.

So even without oil rig, chance of
  extinction is 10% + 70% = 80%. Can’t
  do anything about what’s happening in
  winter migration area.

Suppose their calculations say that
  building will only increase chances of
  extinction to 81%.
Assuming that the data are accurate, the
  cost/benefit ratio in terms of the bird
  hardly changes with or without
  construction.

(Of course, here no direct economic
  value is assigned to the bird, but for
  other aspects of the environment there
  could be).
5.1 Two views of the impacts of a hypothetical project
3) Conversion to monetary terms:

Above, no dollar value assigned, and
  wouldn’t matter much anyway.

Usually though, need some uniform
  scale (usually $$$).

Construction costs are “market goods”
  (can be bought).
Some environmental factors are also
  marketable.
e.g., fishing, hunting etc.

Many environmental factors are
 “nonmarket” but valuable: reduced
 pollution, ecotourism (actually does
 have some market value), role of
 species in ecosystem that has
 measurable effect on humans, etc.

But can’t (usually) put a price on
  enjoying nature.
But, at least a CBA provides some
  baseline that forces the government to
  consider local economy and other
  concerns, and the environment.

Assessing value of biodiversity: use vs.
  non-use values.

Use: fisheries, timber, etc.; indirect use:
  clean water, shade prevents drying =
  long and short-term ecosystem
  services.
5.3 Total economic value of wetlands is composed of direct, indirect,
option, and nonuse values
Ecosystem services: Indirect use; often
  hard to quantify.

e.g., decomposition of organic wastes:
  about 38 billion tons created per year:
  breakdown by microbes, fungi etc.
  frees space, recycles nutrients,
  prevents disease/pollution.

Estimate of total value of ecosystem
  services to US per year: $33 trillion
  (especially wetlands).
5.4 Nonmarket values of
forest can be substantial
Non-use:

No simple good-to-consumer
  relationship; factors such as pleasure
  people get from nature = loss of utility;
  bequest value (passing on to future
  generations).

Add up total use and nonuse values:
  Total Economic Value (TEV).
Historically, nonuse values have been a
  small (or no) part of TEV; now
  becoming a bigger part of the
  equation.

Various ways to convert value of
  biodiversity to money terms.
Substitute value: e.g., how much would it
  cost to buy local vs. foreign wood?

Product function approach: Suppose that
  a change in environment causes a
  measurable change in productivity
  (e.g., desertification).

Opportunity cost: Alternate use value
 (e.g., national park value vs. logging).
Implicit market techniques:

How people act reflects the value that
  they place on the environment: where
  they live, vacation, etc.

“Hedonic” pricing methods: Sum total
  value of a product (or ecosystem)
  based on all individual factors, and
  how does it change if one or more of
  the factors is varied?
One measure: property values. How
  much will people pay to live in a fairly
  natural area vs. a cleared one? (a
  hedonic variable).
But often difficult to assess all the
  nonuse factors.

Travel cost: How far, and how much will
  people pay, to visit, say, a national
  park?
What is a species or ecosystem worth?

Contingent valuation approach: How
  much will people pay to conserve
  something (WTP = “Willingness to
  pay”).

Simulate a market for the species or
  ecosystem: how much will people pay
  to protect it? Depends on the
  individual and the population.
Consider Sri Lankan Sinharaja Forest:
  Estimated TEV via WTP approach.

-Villagers on edge of forest
-More distant rural people
-Sri Lankan urbanites
-UK urbanites

In absolute $$$, richer people would pay
   more to preserve. But relative to
   income, villagers who use the
   resource would pay the most.
  Box 5.1 Table A




WTP = “Willingness to pay?”
5.4 Nonmarket values of
forest can be substantial
How good is the average person at
  estimating value of an ecosystem or
  species?

Survey of coastal and mountain areas in
  Scotland: Scientists assessed species
  richness/diversity, rarity, distributions;
  then ranked in terms of conservation
  priority.

Then asked the public their WTP for
  each area -- very different result.
So -- have to balance conservation
  priorities and satisfying the public, who
  may value the resources differently.
Recall CBA:

1) Project definition
2) Impact classification
3) Conversion to monetary terms

4) Project assessment: Assemble all the
  information in the context of the above,
  and also consider the time frame.
Next table: Developer builds factory,
  breaks even in 10 years, and makes
  more money in the long term.

But -- Environmental cost is high: habitat
  destruction, pollution -- undesirable
  from conservation standpoint.

Politicians look at other costs/benefits,
  including effect on local economy as
  well as environment.
Short term: Economic benefits outweigh
  environmental costs.

But in the long term, environmental costs
  become greater.

Of course, this may not matter so much
  to a politician.

And consider discount rates -- actual $$$
  depreciate, whereas environmental
  value may be more constant.
5.5 Example of how the net present value of a sum varies over 100 years
Last step of CBA:

5) Sensitivity analysis: Given cost/benefit
  estimates, how sensitive are these
  estimates to changes in parameters:
  e.g., product demand, need for clean
  water. Can run simulations under
  different scenarios.
Problems with CBA:

-Uncertainty of estimates themselves
  (especially environmental).

-Defining bounds of inquiry.

- Who benefits -- is it really “all” of
  society?
Case 5.1 (A) The protected area network in Uganda in the mid-1990s
Case 5.1 Table C
Other approaches:

Environmental Impact Assessment:
  Focus on probable environmental
  results of project, either as part of CBA
  or separately.

Often not done very well.
Risk assessment/management:

What is the chance of a hazard; negative
 results. What is the potential
 magnitude of a negative result?

Can express mathematically; e.g., wind
  farm (electricity): may kill birds, bats,
  etc.; depending on level of risk, may
  cancel or relocate.
Many kinds of risk assessment, e.g., use
 of pesticides; intensity of fishing.

Once probability of a given risk is
 calculated, have to decide what’s
 acceptable.

e.g., if harvest of 100 tons of fish/year is
  allowed but this has a 1% chance of
  destroying the fishery, is it worth it?
How does the public perceive the risk?

Similar to a CBA; some things are worth
  more risk than others; e.g., if only a
  few fishermen and not a really in-
  demand species, reduced catch may
  be OK; if this is a main food source for
  a growing population, may be willing to
  take more risk.
Precautionary principle: Despite risk
  assessment, can’t predict all negative
  outcomes (e.g., a medication that
  seems safe may cause cancer later in
  life; effects of DDT on things like bird
  eggshells only discovered after years
  of use and the chemical had gone up
  the food chain).
But how many things to test?

Good example: GMOs (genetically
 modified organisms).

Many people are afraid of these, and
 there are legitimate concerns.

But many of these could have real
  environmental benefits; e.g., drought
  resistance in crops.
So is it better to do nothing? Of course
  not -- but have to assess how much
  testing to do.

Multi-criterion analysis: Pull together
 info. from various kinds of analysis
 (sort of a “meta-analysis”) to find
 optimal solution.
e.g., what approach is most likely to
  preserve habitat, minimize pollution,
  etc.; environmental vs. socioeconomic
  benefits, for example.
5.6 Application of
multi-criteria analysis
Nonsubstitutability: What features of the
 environment absolutely have to be
 protected (for practical or philosophical
 reasons)?

-Ozone layer
-Global warming

“Critical natural capital”.
Are some species or ecosystems better
  or more important than others?

Don’t know how many fit in -- so here the
 precautionary principle comes into
 play.
“Rivet hypothesis” (not in text): Suppose
  that for a plane, losing one rivet
  doesn’t matter much, but
  two...three...ten... (and what part of the
  plane may matter too).

At some point the whole thing falls apart.
How many “rivets” (species/ecosystems)
  can we afford to lose?

Critical reserves: try to save everything
  in a given area -- but practical?

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:2
posted:3/10/2013
language:English
pages:328