FORUM is intended for new ideas or new ways of interpreting existing information. It
provides a chance for suggesting hypotheses and for challenging current thinking on
ecological issues. A lighter prose, designed to attract readers, will be permitted. Formal
research reports, albeit short, will not be accepted, and all contributions should be concise
with a relatively short list of references. A summary is not required.
The role of conservation in expanding biodiversity research
Diane S. Sri7asta7a, Dept of Zoology, Uni7. of British Columbia, 6270 Uni7ersity Bl7d, Vancou7er, B.C. Canada
V6T 1Z4 ([email protected]).
It has been suggested that current reductions in global biodiversity may impair the functioning of ecosystems. This biodiversityecosystem function (BD-EF) hypothesis represents a new avenue
of ecological research originating from conservation concerns.
However, the subsequent evolution of BD-EF research has
reflected academic concerns more than conservation priorities. I
suggest three questions for BD-EF research, which would benefit
both ecological theory and conservation. (1) Is biodiversity the
main driver of ecosystem function? Several experiments show
that biodiversity loss is a minor link between habitat change and
ecosystem function. (2) How will extinction patterns change
BD-EF relationships? Biased extinctions may have additional
impacts on ecosystem function, which can be deduced by comparison with random-loss models. (3) Will conserving regional
biodiversity conserve local ecosystem function? The answer to
this question may differ between saturated and unsaturated
communities, and may depend on whether the magnitude or
stability of ecosystem function is measured.
The world is currently experiencing exceptionally high
rates of species extinctions, largely because of human
activity (Lawton and May 1995). While the loss of
biodiversity is in itself a tragedy, there may be more
practical consequences for humankind. Approximately
a decade ago, several ecologists proposed that reductions in current species diversity would lead to reductions in the functioning of ecosystems; that is, in the
biogeochemical processes carried out by the Earth’s
biota (di Castri and Younes 1990, Ehrlich and Wilson
1991, Lubchenco et al. 1991, Walker 1992, Schulze and
Mooney 1993). The motivation for this biodiversityecosystem function (‘‘BD-EF’’) hypothesis is clearly
rooted in conservation concerns. For example, the Ecological Society of America, in launching its Sustainable
Biosphere Initiative, urged researchers to examine biodiversity effects on ecosystem function because ‘‘ecologists are increasingly asked to justify the benefits of
biological diversity compared to the human benefits
that might be derived from economic development’’
(Lubchenco et al. 1991, p. 390).
The role of conservation in formulating the BD-EF
hypothesis is a reversal of the usual relationship beOIKOS 98:2 (2002)
tween community ecology and conservation. Traditionally, conservation biologists have been quick to adapt
general ecological theories to particular applied issues.
A classic example is the application of island biogeography theory to the design of nature reserves. Community ecologists, by contrast, have been slow to convert
conservation issues into new areas of ecological theory.
Given the importance of conservation for the origin
of the BD-EF hypothesis, one would imagine that as
BD-EF research evolved, it would expand along lines
relevant to conservation. This has been only partially
true. In the last decade, BD-EF research has expanded
in three main areas: methodology, measures of ecosystem function, and modeling of mechanisms. Vigorous
debates around experimental methodology, especially
regarding spurious results, have resulted in increasingly
sophisticated experiments and analysis (Huston 1997,
Loreau 1998a, 1998b, Wardle 1999, Huston et al. 2000).
The original rather vague concept of reduced functioning has been replaced by a more precise understanding
of the key components of ecosystem functioning: magnitude, resilience and resistance to disturbance, constancy in space and time, and resistance to invasions by
exotic species (Case 1990, Tilman 1996, Doak et al.
1998, Naeem 1998). The theoretical foundation for the
BD-EF hypothesis, once meager, has been buttressed
and expanded by recent modeling studies (Tilman et al.
1997, 1998, Ives et al. 1999, Loreau 2000, McCann
2000). While all of these areas are academically important, and indeed essential for the development of the
discipline, few have direct relevance to conservation (a
notable exception being interest in stability benefits of
biodiversity: Schwartz et al. 2000). Nor has BD-EF
research crossed over to any degree into the conservation literature (as evidenced by the concentration of this
research in academic ecology journals: see reference list
of this paper). The purpose of this paper is to suggest
several new directions for BD-EF research that would
make it more useful to conservation. However, making
BD-EF research more useful for conservation need not
come at the expense of ecological generality. Indeed,
the research questions outlined in this paper also point
the way to how existing ecological theory can be integrated into a BD-EF framework. I begin by discussing
how applied goals can be accommodated within academic ecology.
Are academic and applied goals compatible
in biodiversity research?
Conservation and community ecology have different
goals for biodiversity research (Takacs 1996). The ecological or academic goal for BD-EF research is to
determine the presence of biodiversity effects on ecosystem function. The conservation or applied goal for
BD-EF research is to determine the importance and
predict the likelihood of such a biodiversity effect. These
academic and applied goals are occasionally in conflict,
as illustrated by a recent debate in the literature. The
‘‘sampling effect’’ debate can be summarized briefly as
follows. In many BD-EF experiments, communities of
differing diversity are randomly assembled from a species pool. As diversity increases, the probability of any
given species occurring in the community also increases.
If one particular species has a disproportionate effect
on the ecosystem function being measured, any increase
in ecosystem function with diversity could simply reflect
the increasing the probability of including this species
(a ‘‘sampling’’ or ‘‘selection probability’’ effect) rather
than a true biodiversity effect (Aarssen 1997, Huston
1997, Wardle 1999). This is an academic argument,
which questions the presence of a biodiversity effect
once external, non-ecological mechanisms (sampling
probability) are accounted for. The most common
counter-argument, by contrast, is based in applied concerns and is as follows. Even if the underlying mechanism behind BD-EF correlations turns out to be
probability rather than ecology, this does not invalidate
the conclusion that declining biodiversity will affect
ecosystem function. The experiments mimic how humankind is reducing biodiversity: randomly, without
regard to any particular species’ effect on ecosystem
function (Lawton et al. 1998, van der Heijden 1999,
Hector et al. 2000). Although sampling probability is
not an academically exciting mechanism, it may be an
important mechanism in the real world.
The ‘‘sampling effect’’ debate also illustrates how
academic and applied goals can be reconciled. Recent
methods allow ecologists to assess the contribution of
sampling effects to BD-EF relationships (Hector 1998,
Loreau 1998a, 1998b). These methods allow academic
goals to be met, as correcting for sampling effects
allows us to search for more ecologically-interesting
processes. It also satisfies applied goals, as it allows us
to assess the relative contribution of different mechanisms to an overall biodiversity effect. Once these goals
are recognized as distinct from each other, conflict
disappears (as has recently been witnessed: Hughes and
Petchey 2001, Loreau et al. 2001).
I now turn to several specific examples of how BDEF research could be expanded in ways that satisfy
both academic and applied goals. In these examples,
academic goals are met by linking current BD-EF
theory with other ecological theory, such as species
saturation theory or life history correlates of extinction
risk. Applied goals are met by asking questions about
the importance of biodiversity effects in the context of
other environmental changes, and by linking local biodiversity effects with regional conservation policies.
Is biodiversity the main driver of ecosystem
Many BD-EF experiments have shown that biodiversity
has some effect on ecosystem function (reviewed by
Schwartz et al. 2000). It is not clear, however, whether
this biodiversity effect is important in comparison to
the more direct effects of habitat change on ecosystem
function (Hughes and Petchey 2001).
Habitat change is generally agreed to be the primary
reason for the current wave of species extinctions (Diamond 1989, Lawton and May 1995). Close on its heels
are species invasions and over-exploitation (Diamond
1989). Climate change is poised to become a fourth
major cause of contemporary extinctions (Sala et al.
2000). Clearly, the only way to prevent further biodiversity loss is by ameliorating these factors. At some
point, therefore, BD-EF research needs to explicitly
examine whether changes in human action (e.g. conserving intact habitat) intended to protect biodiversity
will also preserve ecosystem function.
The links between habitat change, biodiversity loss
and ecosystem function are surprisingly complex. Consider the case of habitat fragmentation. There would
appear to be a solid case for reducing habitat fragmentation if it is conclusively shown both that habitat
fragmentation leads to lower local diversity, and that
lower local diversity leads to reduced ecosystem function. However, this argument is not logically complete
as it assumes that there is no direct (i.e., non-biodiversity) effect of habitat fragmentation on ecosystem function (Fig. 1). This is unlikely to be the case. Habitat
fragmentation can cause a variety of edaphic changes,
such as increased wind exposure, higher evaporation
rates, and greater light penetration (Murcia 1995), all of
which could directly affect any number of ecosystem
functions. The key questions will be the sign and magnitude of such direct effects of habitat change on
ecosystem function, relative to biodiversity effects on
ecosystem function.
OIKOS 98:2 (2002)
There are a number of experiments which suggest
that direct effects between habitat changes and ecosystem function can be of greater magnitude than biodiversity effects on ecosystem function, and even
counteract the effects of biodiversity loss. Petchey et al.
(1999) used aquatic micro-organisms in experimental
microcosms to examine the effect of environmental
warming on primary production. Warming increased
primary production in two ways, directly through increasing physiological rates, and indirectly through
changing biodiversity and trophic structure (Fig. 1b).
The gradual 14°C increase in temperature roughly doubled primary production; about 2/3 of this increase
could be attributed simply to increased physiological
rates (using standard Q10 equations), leaving up to 1/3
due to food-web changes (e.g. increased species extinctions and fewer trophic levels) (O. Petchey, pers.
comm.). Other treatments in this experiment confirm
that reduced biodiversity in this system can lead to
increased primary production, independent of temperature change. Wardle et al. (1997) present observational
evidence that several ecosystem functions (notably decomposition and nitrogen mineralization) are positively
correlated with island area. At first glance, one might
expect that this correlation could be explained by biodiversity effects alone; after all, islands are well-known to
have strong species-area relationships. In this case,
however, these island biogeography effects are com-
pletely overwhelmed by a positive correlation between
island area and fire frequency (from lightning strikes).
The higher fire frequency on large islands not only
depresses species diversity below that on small islands
but, more importantly, radically shifts the composition
of the leaf litter to faster-decomposing species (Fig. 1c).
Finally, in an experiment described by Gonzalez and
Chaneton (2002), fragmentation of moss habitat reduced micro-arthropod productivity, but only 10% of
this ecosystem function effect was due to reduced species richness; the remainder was due to decreases in
total abundance through dispersal limitation (Fig. 1d).
All of these studies allowed biodiversity to be naturally affected by an extrinsic factor (simulated global
warming in Petchey et al. 1999; habitat area in Wardle
et al. 1997; fragmentation in Gonzalez et al. 1998). This
particular design feature allowed biodiversity effects on
ecosystem function to be compared with direct effects
of the factor. By contrast, most BD-EF studies reduce
biodiversity independently of any extrinsic factor. Although the latter approach is valuable in determining
the presence of biodiversity effect on ecosystem function, it does not allow us to determine the importance of
that biodiversity effect relative to other effects. These
two approaches should thus be seen as complementary,
and it is particularly valuable when both approaches
are integrated to a single study (as in Petchey et al.
1999, Levine 2000). Ultimately, to fully understand the
Fig. 1. (a) Habitat changes can
affect ecosystem functioning both
directly and indirectly via
biodiversity effects. Specific
examples of this phenomenon are
evident in: (b) aquatic microbial
communities experimentally
manipulated by Petchey et al.
(1999), (c) boreal island vegetation
studied by Wardle et al. (1997),
and (d) micro-arthropods in moss
in an experiment described in
Gonzalez and Chaneton (2002).
The width of each arrow
approximates the relative strength
of the effect as determined by the
OIKOS 98:2 (2002)
repercussions of human activity on the functioning of
ecosystems, we need to understand the importance of
all intermediate pathways, including but not restricted
to biodiversity loss.
How will extinction patterns change BD-EF
All species may go extinct, but some species are more
likely to go extinct than others. Particularly vulnerable
to extinction are large-bodied, K-selected species which
occupy high trophic positions in food webs and occur
at low abundance (Lawton and May 1995). For example, there is clear evidence that for at least the last 30
years, humans have been consistently ‘‘fishing down
food webs’’; that is, harvesting fish species in top
trophic levels under they reach economic extinction,
then moving to species at progressively lower trophic
levels (Pauly et al. 1998). By contrast, many BD-EF
experiments assume random loss of species, and have
been criticized as not representing real extinction biases
between species (Wardle 1999, Griffiths et al. 2000).
The counter-argument is that, if experiments were designed with some species being preferentially lost over
others, then any general effect of biodiversity on
ecosystem function would be confounded by a systematic change in species composition (Huston 1997, Lawton et al. 1998, Hector et al. 2000). In many ways, this
is a debate reminiscent of the ‘‘sampling effect’’ debate:
a conflict between pragmatic, conservation concerns
and a more academic viewpoint striving for unbiased
effects. There is a middle way to view this debate.
Random-loss experiments could provide the critical
null-model for examining the effects of biased
For example, suppose we wanted to look at the effect
of rarity on modifying BD-EF relationships. If we
assumed that the probability of a species’ extinction
exponentially declines with population size, we could
use information on the natural abundances of species to
estimate the relative probability of each species going
extinct (Fig. 2). Experimental communities of varying
diversities could then be assembled using these speciesspecific probabilities: these communities represent a
biased-extinction scenario. At the same time, we could
assemble a second series of communities that would
represent a true random-loss scenario: each species
would have an equal probability of inclusion in a
community. By comparing how ecosystem function is
affected by biodiversity loss in each scenario, we could
separate the effect of losing biodiversity per se from the
effect of losing rare species preferentially. In other
words, we ask the question whether the absolute effect
of biodiversity loss (random-loss scenario) differs from
the likely effect of biodiversity loss (biased-loss sce354
nario). Any attempts to translate BD-EF findings into
public policy will need to consider such disparities
between theory and practice.
One can even go a step further here, and allow
diversity to be naturally decreased in a BD-EF experiment by a habitat change, such as fertilization (Tilman
1996), warming (Petchey et al. 1999) or pollution
(Griffiths et al. 2000), as discussed earlier. This will
cause biased extinctions. Such an approach, when coupled with appropriate random-loss and biased-loss experiments run at the same time, could potentially tease
apart (1) the biodiversity-caused effects on ecosystem
function from more direct effects of the perturbation,
and (2) the contributions of biased loss versus random
loss to this biodiversity effect.
Will conserving regional biodiversity conserve
local ecosystem function?
Biodiversity-ecosystem function experiments are typically carried out on small spatial scales, as small as a
single Petri dish for a protist community, or a 1 ×1 m
plot for a plant community. Such experiments link
changes in local diversity with changes in local function, where ‘‘local’’ is defined as the scale at which
ecological interactions between species occur (Cornell
and Lawton 1992, Ricklefs and Schluter 1993).
In the real world, however, biodiversity loss is occurring at a variety of spatial scales: at the small scale of
local communities, at the larger scale of regional species
pools, and ultimately at the scale of the entire globe.
There are valid conservation concerns for biodiversity
loss at all of these scales. However, as the scale of
biodiversity loss increases – from local to regional to
global extinctions – concern generally mounts. Most
conservation guidelines prioritize species at risk of
large-scale extinction over those at risk of local extinction (e.g. International Union for the Conservation of
Nature guidelines). In the few countries which have
formal protection for endangered species, the legislation
is enacted by federal or provincial governments and
concerns only nationally- or provincially-threatened
species (Ray and Ginsberg 1999). At some stage, therefore, BD-EF research will need to examine the effects
of biodiversity loss at these larger scales. The next
challenge for ecologists is to link regional-level extinctions with changes in ecosystem function.
There are several steps in the pathway between regional biodiversity and local ecosystem function (Fig.
3a). First, changes in regional biodiversity may or may
not affect local biodiversity. Secondly, changes in local
biodiversity may or may not affect local ecosystem
function (the more familiar question). I will examined
each of these links in turn, and argue that both links
depend on whether communities are ‘‘saturated’’ or
‘‘unsaturated’’ with species.
OIKOS 98:2 (2002)
Fig. 2. An experimental
design allowing the functional
consequences of biased
species loss to be compared
to random species loss. To
generate the biased-loss
extinction probabilities, I
simply assumed the
probability of species survival
was exponentially related to
its population size; more
complex models could be
used here. The relative
probability of excluding a
species from an experimental
community would be
equivalent to its relative
probability of extinction.
Let us begin with the first question: whether loss of
regional diversity will necessarily result in loss of local
diversity. Obviously, if any species goes regionally extinct, it is lost from all localities in that region, leading
to an instantaneous decrease in local diversity. However, this reduction in local diversity may simply be a
transient effect (Grime 1998). Species from other locations in the region may well invade the depauperate
communities, returning local richness to its previous
level. This replacement process will tend to decouple
local diversity from regional diversity. In fact, this
replacement process could allow substantial reductions
in regional diversity with no effect on local diversity, as
OIKOS 98:2 (2002)
long as species remain in the species pool with the right
attributes to quickly fill each niche as it becomes vacant. At some point, however, the pool will become
exhausted of suitable replacement species, and local
diversity will begin to fall with regional diversity (Fig.
The effect of regional extinctions on local diversity
therefore depends on where the original community
falls on this local-regional richness trajectory. Ecologists already distinguish communities according to such
effects of regional diversity on local diversity: ‘‘unsaturated’’ communities exhibit strong effects of regional
diversity on local diversity whereas the local diversity of
Fig. 3. (a) Changes in
regional biodiversity are
predicted to affect local
ecosystem function via effects
on local biodiversity. The
strength of these links will
differ between (b) saturated
and (c) unsaturated
‘‘saturated’’ communities is relatively independent of
variations in regional-level diversity (Fig. 4a). Although
the concept of saturation has traditionally been used in
discussing the accumulation and maintenance of local
diversity, it can also apply to the process of biodiversity
Saturated and unsaturated communities will differ
not only in the effect of regional diversity on local
diversity, but also in the effect of local diversity on
ecosystem function. The latter effect depends on differences in interaction strengths between saturated and
unsaturated communities. Recall that in saturated communities, lost species are quickly replaced with new
species because the species pool generally contains numerous species able to fill any particular niche. Such a
high species: niche ratio is predicted to lead to strong
interspecific competition (Cornell and Lawton 1992,
but see Fox et al. 2000). Therefore, even if local diversity was successfully reduced in a saturated community,
such strong competitive interactions would likely lead
to an increase in the abundance of the remaining
species; i.e. density compensation. This compensatory
effect would diminish any effect of species loss on
ecosystem function (Ruesink and Srivastava 2001).
By contrast, in unsaturated communities, there are
few if any competitors for a given niche (this is why lost
species are generally not replaced). In such weakly-interactive communities, density compensation is not predicted for lost species. Thus, unsaturated/weaklyinteractive communities are likely to show strong effects
of local diversity on ecosystem function (Ruesink and
Srivastava 2001). This conclusion agrees with several
theoretical models for BD-EF relationships. The ‘‘niche
differentiation’’ or ‘‘efficiency’’ effect (Tilman et al.
1997, Loreau 1998a, 1998b, 2000) is predicted to result
in strong BD-EF correlations when niche overlap is
minimal between species, as in unsaturated, weakly-interactive communities.
If we combine all the above effects, saturated communities should exhibit minimal effects of reduced re356
gional richness on local ecosystem function; not only is
regional richness decoupled from local richness but
local richness itself is not linked to ecosystem function
Fig. 4. (a) Loss of regional diversity may have minimal affect
on local diversity as long as communities are saturated with
species. When communities become unsaturated with species,
local diversity will fall, reducing (b) local ecosystem function.
(c) Stability is predicted to show the opposite relationship with
regional diversity. The exact shapes of these relationships are
OIKOS 98:2 (2002)
because of strong interspecific competition (Fig. 3b).
Conversely, in unsaturated communities strong links
from regional richness to local richness through to
ecosystem function should ensure a strong role for
regional richness in determining ecosystem function
(Fig. 3c). Note that here I am considering ecosystem
consequences of extinctions in a single region, rather
than simply variance in richness between regions; in the
latter case, ecosystem functions likely become drivers
not responders of diversity (Huston 1997, Hector et al.
1999, Loreau 2000). Given that the consequences of
reduced regional biodiversity for ecosystem function
will differ between saturated and unsaturated communities, it seems imperative that ecologists begin to distinguish between these types of communities in both
BD-EF theory and experiments. There are a number of
ways to test for saturation, including biogeographical
comparisons, evidence of competition and density compensation, and monitoring invasion effects (Ricklefs
and Schluter 1993, Srivastava 1999).
It should be clear by now that BD-EF experiments
carried out at the local scale cannot simply be ‘‘scaledup’’ to the regional scale. There is no straightforward
link between regional extinctions and changes in
ecosystem function; rather, regional extinctions must be
translated to biodiversity changes at the local scale
before ecosystem function can be altered. Nor can we
ignore the importance of regional-level changes in biodiversity. Ecologists will do conservation a great disservice if we do not make clear that the effects of
biodiversity loss depend on scale. Some forestry companies are already describing clear cuts as having biodiversity value because they contain more species per
hectare than the original forest. If we only value biodiversity at the local scale, the forest companies may be
right; however, if we view clear cuts as replacing regionally-rare old-growth species with common secondarygrowth species, the logic no longer holds. There is also
a superb opportunity for ecological theory here, to
construct a truly multi-scale theory of biodiversity
which integrates regional and landscape-level processes
with the effects of small-scale interactions between
Regional biodiversity and ecosystem stability
Biodiversity may be critical not only for maintaining
the current magnitude of ecosystem function, but also
its stability over time. Stability has much value for
human society. Unexpected drops or surges in ecological processes can destabilize local economies, disrupt
social systems, and discourage investment. Recent examples of catastrophic ecosystem failure include the
crash of Eastern Canada’s cod stocks, landslides in
deforested areas of Central America, and the drying-up
OIKOS 98:2 (2002)
of the Aral Sea. Regional biodiversity can be linked to
the stability of ecosystem function in two ways: by
providing an adequate species pool for strongly-interactive communities, and by providing a reservoir of potentially important species for the future.
Most hypotheses which link biodiversity to the stability of ecosystem function require strong interspecific
interactions (Ruesink and Srivastava 2001). To give one
example, the competition hypothesis (Tilman 1996,
Naeem and Li 1997) suggests that reduction in the
abundance of any species in high diversity communities
is likely to be matched by an increase in the abundance
of a competitor. Thus strong competitive interactions
result in dampened variation in total abundance, and
hence ecosystem function. (Low diversity communities,
by contrast, are less likely to contain appropriate competitors, so have greater variation). This hypothesis has
been supported by some models (Tilman et al. 1997,
1998) but not others (Ives et al. 1999, Loreau and
Behara 1999). According to the logic of the previous
section, we would expect saturated communities to
have the strong competitive interactions required for
stability. Ironically, in saturated communities, local
biodiversity may be strongly correlated with the stability of ecosystem function but weakly correlated with its
magnitude (see also Ruesink and Srivastava 2001). As
species go regionally extinct, local communities will
tend to be less and less saturated with species, and have
increasingly less stable ecosystem functions (Fig. 4c).
High regional richness may also stabilize ecosystem
functions in periods of dramatic environmental change.
The ‘‘insurance hypothesis’’ (Tilman 1996, Yachi and
Loreau 1999) proposes that communities depauperate
in species are less likely to contain a species critical to
maintaining ecosystem function under different environmental conditions in the future. If we expand this
argument to the regional scale, then regions depauperate in species are less likely to be able to contain a
species critical for maintaining local ecosystems in the
future. Thus, even if reductions in regional richness do
not affect local richness today (for example, because
local communities are saturated with species), there
may be consequences for future ecosystem function.
This interpretation of the insurance hypothesis confers
a local stability benefit to regional scale biodiversity,
and draws attention to the value of beta diversity, that
is, spatial differences in species composition.
How can regional diversity be manipulated?
Regional extinctions need to be explicitly incorporated
into experiments and models. Of course, it is very
difficult in practice to manipulate biodiversity at the
regional level. This scale constraint could be circumvented in a variety of ways. First of all, changes in
regional diversity can be simulated by supplying local
experimental plots with propagules representing different numbers of species (i.e., different ‘‘species pools’’),
not all of which will establish. In some ways, this
experiment is already being conducted by plant ecologists. Recent experiments as part of the BIODEPTH
project (e.g. Hector et al. 1999) and Cedar Creek
program (e.g. Tilman et al. 1996) involved seeding plots
with a constant mass of seeds representing different
numbers of species (different species pools), only some
of which survived in the plots (different local diversities). In both cases, there were strong links between
planned and actual local diversity and between planned
diversity and ecosystem function, implying strong effects of the species pool on ecosystem function. In a
similar vein, Levine (2000) described how landscapelevel variation in plant propagule supply affected native
plant diversity on river tussocks, which in turn affected
(in controlled experiments) the invasion resistance of
each tussock to exotic species.
A second approach is to use small-bodied species in
appropriately small-sized landscapes to make manipulations of regional diversity manageable. Ideal candidates
for this type of experiment are protist communities and
micro-arthropod communities. In protist communities,
correlations between local and regional diversity have
been shown (Fox et al. 2000), and between local diversity and ecosystem function (Naeem and Li 1997,
Petchey et al. 1999), but not yet between true regional
diversity and ecosystem function. In the microarthropod communities on moss, secondary productivity appears to be linked to local richness (Gonzalez and
Chaneton 2002). We (A. Gonzalez and D. Srivastava)
are now conducting experiments in which we manipulate the regional richness of moss micro-arthropods by
changing the size of large patches of moss (‘‘regions’’),
to examine the effect first on local diversity and then on
ecosystem function.
A third approach is to use various ‘‘natural experiments’’, such as are provided by recent regional extinctions or declines (for example, bird species on Guam),
species invasions (Vitousek 1990), or biogeographic effects on diversity (Feinsinger et al. 1982, Wardle et al.
1997) . I provide examples of each of these in turn. The
rapid population decline of Aleutian island sea otters
triggered a trophic cascade resulting in an order of
magnitude decline in kelp productivity, an important
marine ecosystem function (Estes et al., 1998). By contrast, the sudden loss of North American chestnut trees
due to blight had remarkably little effect on the Lepidopteran species which formerly fed upon this tree:
only 7 of 56 Lepidopteran species went extinct (Pimm
1991). In6ading species occasionally have dramatic effects on ecosystem functions. For example, invasion of
Hawaii by the exotic shrub Myrica faya has quadrupled
nitrogen inputs in early-succession ecosystems (Vitousek 1990). Many other invasive species, by contrast,
may have little effect on ecosystems (Vitousek 1990). It
is also still not clear if the ecosystem effects of adding
an exotic species are expected to be symmetrical to the
effects of losing a native species. Island biogeographic
effects are particularly well-documented for the
Caribbean islands. The island of Tobago has a smaller
regional species pool of hummingbirds (5 species) than
the island of Trinidad (16 species), presumably because
of its smaller size and longer period of isolation. This
difference in regional diversity appears to have resulted
in lower (and more variable) pollination rates of many
flower species on Tobago relative to Trinidad
(Feinsinger et al. 1982).
The origin of the BD-EF hypothesis represents a
tremendous opportunity to link the disciplines of community ecology and conservation. This is, however, not
a natural partnership. Community ecologists are
trained to ask questions about mechanism and, to do
so, abstract communities from complicating external
factors. This makes ecologists excellent detectives, but
poor forecasters. Without knowing the relative importance of these ‘‘externalities’’ it is difficult to predict
how important a particular mechanism (in this case
biodiversity loss) will be in the future. Unfortunately,
conservation requires precisely this type of forecasting
ability: policymakers generally decide on conservation
measures by weighing the likelihood of success against
the costs. I have outlined how BD-EF research could
incorporate three ‘‘externalities’’: causes of extinctions,
patterns of extinctions, and the spatial scale of extinctions. Each of these directions would make biodiversity
research more applicable to conservation. Furthermore,
each direction is made possible by integrating existing
ecological theory into the current BD-EF framework,
creating a more comprehensive theory of biodiversity.
Both ecology and conservation would benefit by such
an expansion of the scope of BD-EF research.
Biodiversity is a challenging phenomenon to study,
for it reflects the confluence of ecological, evolutionary
and anthropogenic processes, many of which operate
on different spatial and temporal scales. Some of the
most exciting recent advances in biodiversity research
reflect links made between these different processes and
scales (Ricklefs and Schluter 1993, Gaston 2000,
Schluter 2000). Ecologists are now on the brink of
linking human activity with the functional effects of
biodiversity. Our success in this endeavor will depend
on our creativity in translating conservation concerns
into testable ecological hypotheses.
Acknowledgements – The manuscript was improved by comments from several ecologists at the University of British
Columbia, particularly Ramona deGraaf, Chris Lortie, Tom
Oakey, Jordan Rosenfeld, Dolph Schluter, Susan Shirley, and
OIKOS 98:2 (2002)
Roy Turkington, as well as two anonymous reviewers. Thanks
to Enrique Chaneton, Andy Gonzalez and Owen Petchey for
access to unpublished manuscripts and calculations.
Aarssen, L. W. 1997. High productivity in grassland ecosystems: effected by species diversity or productive species? –
Oikos 80: 183 –184.
Case, T. J. 1990. Proc. Natl. Acad. Sci. USA 87: 9610.
Cornell, H. V. and Lawton, J. H. 1992. Species interactions,
local and regional processes, and limits to the richness of
ecological communities: a theoretical perspective. – J.
Anim. Ecol. 61: 1 –12.
di Castri, F. and Younes, T. 1990. Ecosystem function of
biological diversity. Biology International, Special Issue
Number 22.
Diamond, J. M. 1989. The present, past and future of humancaused extinctions. – Phil. Trans. Roy. Soc. B. 325: 469 –
Doak, D. F., Bigger, D., Harding, E. K. et al. 1998. The
statistical inevitability of stability-diversity relationships in
community ecology. – Am. Nat. 151: 264 –276.
Ehrlich, P. R. and Wilson, E. O. 1991. Biodiversity studies:
science and policy. – Science 253: 758 –761.
Estes, J. A., Tinker, M. T., Williams, T. M. and Doak, D. F.
1998. Killer whale predation on sea otters linking oceanic
and nearshore ecosystems. – Science 282: 473 –476.
Feinsinger, P., Wolfe, J. A. and Swarm, L. A. 1982. Island
ecology: reduced hummingbird diversity and the pollination biology of plants, Trinidad and Tobago, West Indies.
– Ecology 63: 494 –506.
Fox, J. W., McGrady-Steed, J. and Petchey, O. L. 2000.
Testing for local species saturation with nonindependent
regional species pools. – Ecol. Lett. 3: 198 –206.
Gaston, K. J. 2000. Global patterns in biodiversity. – Nature
405: 220 – 227.
Gonzalez, A. and Chaneton, E. J. 2002. Heterotroph species
extinction, abundance and biomass dynamics in an experimentally fragmented microecosystem. – J. Anim. Ecol. in
Gonzalez, A., Lawton, J. H., Gilbert, F. S. et al. 1998.
Metapopulations dynamics, abundance, and distribution in
a macroecosystem. – Science 281: 2045 –2047.
Griffiths, B. S., Ritz, K., Bardgett, R. D. et al. 2000. Ecosystem response of pasture soil communities to fumigationinduced microbial diversity reductions: an examination of
the biodiversity-ecosystem function relationship. – Oikos
90: 279 – 294.
Grime, J. P. 1998. Benefits of plant diversity to ecosystems:
immediate, filter and founder effects. – J. Ecol. 86: 902 –
Hector, A. 1998. The effect of diversity on productivity:
detecting the role of species complementarity. – Oikos 82:
597 – 599.
Hector, A., Schmid, B., Beierkuhnlein, C. et al. 1999. Plant
diversity and productivity experiments in European grasslands. – Science 286: 1123 –1127.
Hector, A., Schmid, B., Beierkuhnlein, C. et al. 2000. Response. – Science 289: 1255a.
Hughes, J. B. and Petchey, O. L. 2001. Merging perspectives
on biodiversity and ecosystem functioning. – Trends Ecol.
Evol. 16: 222 –223.
Huston, M. A. 1997. Hidden treatments in ecological experiments: re-evaluating the ecosystem function of biodiversity.
– Oecologia 110: 449 –460.
Huston, M. A., Aarssen, L. W., Austin, M. P. et al. 2000. No
consistent effect of plant diversity on productivity. – Science 289: 1255a.
Ives, A. R., Gross, K. and Klug, J. L. 1999. Stability and
variability in competitive communities. – Science 286:
542 – 544.
OIKOS 98:2 (2002)
Lawton, J. H. and May, R. M. 1995. Extinction rates. –
Oxford Univ. Press.
Lawton, J. H., Naeem, S., Thompson, L. J. et al. 1998.
Biodiversity and ecosystem function: getting the Ecotron
experiment in its correct context. – Funct. Ecol. 12: 848 –
Levine, J. M. 2000. Species diversity and biological invasions:
relating local process to community pattern. – Science 288:
852 – 854.
Loreau, M. 1998a. Biodiversity and ecosystem functioning: a
mechanistic model. – Proc. Natl. Acad. Sci. USA 95:
5632 – 5636.
Loreau, M. 1998b. Separating sampling and other effects in
biodiversity experiments. – Oikos 82: 600 – 602.
Loreau, M. 2000. Biodiversity and ecosystem functioning. –
Oikos 91: 3 – 17.
Loreau, M. and Behara, N. 1999. Phenotypic diversity and
stability of ecosystem processes. – Theor. Pop. Biol. 56:
29 – 47.
Loreau, M., Naeem, S., Inchausti, P. et al. 2001. Biodiversity
and ecosystem functioning: current knowledge and future
challenges. – Nature 294: 804 – 808.
Lubchenco, J., Olson, A. M., Brubaker, L. B. et al. 1991. The
sustainable biosphere initiative: an ecological research
agenda. – Ecology 72: 371 – 412.
McCann, K. S. 2000. The diversity-stability debate. – Nature
405: 228 – 233.
Murcia, C. 1995. Edge effects in fragmented forests: implications for conservation. – Tree 10: 58 – 62.
Naeem, S. 1998. Species redundancy and ecosystem reliability.
– Cons. Biol. 12: 39 – 45.
Naeem, S. and Li, S. 1997. Biodiversity enhances ecosystem
reliability. – Nature 390: 507 – 509.
Pauly, D., Christensen, V., Dalsgaard, J. et al. 1998. Fishing
down marine food webs. – Science 279: 860 – 863.
Petchey, O. L., McPhearson, P. T., Casey, T. M. and Morin,
P. J. 1999. Environmental warming alters food-web structure and ecosystem function. – Nature 402: 69 – 72.
Pimm, S. L. 1991. The balance of nature? – Univ. of Chicago
Ray, J. C. and Ginsberg, J. R. 1999. Endangered species
legislation beyond the borders of the United States. –
Cons. Biol. 13: 956 – 958.
Ricklefs, R. E. and Schluter, D. 1993. Species diversity in
ecological communities: historical and geographical perspectives. – The Univ. of Chicago Press.
Ruesink, J. L. and Srivastava, D. S. 2001. Numerical and per
capita responses to species loss: mechanisms maintaining
ecosystem function in a community of stream insect detritivores. – Oikos 93: 221 – 234.
Sala, E. O., Chapin III, F. S., Amnesto, J. J. et al. 2000.
Global biodiversity scenarios for the year 2100. – Science
287: 1770 – 1774.
Schluter, D. 2000. The ecology of adaptive radiation. – Oxford Univ. Press.
Schulze, E. D. and Mooney, H. A. 1993. Biodiversity and
ecosystem function. – Springer-Verlag.
Schwartz, M. W., Brigham, C. A., Hoeksema, J. D. et al.
2000. Linking biodiversity to ecosystem function: implications for conservation ecology. – Oecologia 122: 297 – 305.
Srivastava, D. S. 1999. Using local-regional richness plots to
test for species saturation: pitfalls and potentials. – J.
Anim. Ecol. 68: 1 – 16.
Takacs 1996. The idea of biodiversity: philosophies of paradise. – John Hopkins Univ. Press.
Tilman, D. 1996. Biodiversity: population versus ecosystem
stability. – Ecology 77: 350 – 363.
Tilman, D., Lehman, C. L. and Bristow, C. E. 1998. Diversitystability relationships: statistical inevitability or ecology
consequences? – Am. Nat. 151: 278 – 282.
Tilman, D., Lehman, C. L. and Thomson, K. T. 1997. Plant
diversity and ecosystem productivity: theoretical considerations. – Ecology 94: 1857 –1861.
Tilman, D., Wedin, D. and Knops, J. 1996. Productivity and
sustainability influenced by biodiversity in grassland
ecosystems. – Nature 379: 718 –720.
van der Heijden, M. G. A. 1999. ‘Sampling effect’, a problem
in biodiversity manipulation? A reply to David A. Wardle.
– Oikos 87: 408 –409.
Vitousek, P. M. 1990. Biological invasions and ecosystem
processes: towards an integration of population biology
and ecosystem studies. – Oikos 57: 7 –13.
Walker, B. H. 1992. Biodiversity and ecological redundancy. –
Cons. Biol. 6: 18 – 23.
Wardle, D. A. 1999. Is ‘sampling effect’ a problem for experiments investigating biodiversity-ecosystem function relationships? – Oikos 87: 403 – 407.
Wardle, D. A., Zackrisson, O., Hörnberg, G. and Gallet, C.
1997. The influence of island area on ecosystem properties.
– Science 277: 1296 – 1299.
Yachi, S. and Loreau, M. 1999. Biodiversity and ecosystem
productivity in a fluctuating environment: the insurance
hypothesis. – Proc. Natl. Acad. Sci. USA 96: 1463 –
OIKOS 98:2 (2002)
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Related manuals

Download PDF