Diversity - productivity relationships

1
Diversity - productivity relationships
Productivity Rate of energy flow to a system
(averaged over an appropriate time scale).
Typically used by ecologists to denote net rate
at which CO2 is fixed by the ecosystem (Gross
productivity) Diversity Usually means species
richness in this context Productivity is
difficult to measure directly. So, generally some
indirect measure that is correlated to
productivity is used. Examples?
2
Huston (1993) Suggested that exists a negative
relationship between resource availability and
species richness Argues - For plants Under
favorable conditions the strongest competitors
are able to dominate rapidly and completely and
eliminate inferior species through resource
capture (generally light). - Somewhat poorer
conditions, superior competitors are not able to
dominate as rapidly (if at all) and more species
are able to coexist. - Overall pattern therefore
is for highest plant diversity on poor soils and
low plant diversity on the best soils. -
Insect/bird diversity closely related to plant
diversity therefore overall biodiversity highest
in poor sites
3
Huston concluded areas with highest plant
productivity would be those least important to
the conservation of biological diversity Situat
ions where agricultural productivity is
inherently high are situations where the natural
biodiversity of plants would be expected to be
relatively low. Thus preserving a large
proportion of the plant diversity in the world
should not have much social or economic cost
because it does not require a sacrifice of
agriculturally productive land Implication
Biological conservation and agricultural
production are not in conflict. Hustons paper
clearly written with policy makers in mind drew
a lot of criticism from conservationists!
4
Conflict between agriculture and protection of
diversity does exist (or did in Illinois)
Predicting effect of changing productivity on
species richness is also important because of
widespread fertilization by N deposition. (e.g.
converting N poor heathlands to grasslands), and
for the prospects of restoring species rich
communities to abandoned and agriculturally
improved land. Need to understand - The scale
at which diversity varies in relation to
productivity - Whether the pattern that Huston
describes is real and general - The mechanisms
underlying any pattern
5
General pattern described is unimodal or humped
Many examples of plant diversity showing a humped
relationship with some index of productivity (eg
plant biomass, soil P and K concentration, Soil
moisture. (see Morin book p. 333 for more
examples)
Mittelbach et al. (review in Ecology 2001) found
that the unimodal relationship was particularly
common for plant community studies in which
biomass was used as the correlate for
productivity.
6
Also found for other correlates of productivity
(e.g. moisture availability in prairie land
7
and indices of soil fertility (this case a
normalized index of soil P and K concentrations.
(7 data points!)
8
Hump-shape relationships also common in lakes
(Dodson 2000)
Species richness per lake after adjusting for
lake size
Log species richness
Log primary productivity
9
What mechanisms could generate this pattern?
1.Heterogeneity in resource supply
rates Tilman and Pacala (1993) suggest that
Tilmans R model could generate this pattern -
remembering that heterogeneity in 2 potentially
limiting resource supply rates (eg light and
nutrients) can permit multiple species
coexistence. - Argues that highest diversity
will be maintained when greatest heterogeneity in
supply rates of resources exists. This will not
take place when one resource is either very
scarce or abundant. Actual location of highest
predicted diversity will depend on the scale of
resource heterogeneity relative to scale of plant
size. Should be experimentally testable
hypothesis
10
Leibold (1996) Low productivity predators are
too rare to mediate competition among prey and
diversity is low because the best competitors
dominate Intermediate productivity Support
higher predator densities - sufficient to reduce
the abundance of competitively superior (but
predator sensitive) prey and allow coexistence
with inferior (predator resistant) prey High
productivity Predators abundant. Predator
sensitive prey eliminated. Only predator
resistant prey survive. No experimental tests of
this hypothesis? General enough to explain hump
shape in so many communities?
2. Varying intensities of keystone predation
11
3. Assemblage level thinning (Stevens and Carson
1999, Oksanen 1996).
Hump shaped is an artifact of area-based
sampling? As productivity increases so does the
size of individual plants. Space constraints
means that increased size results in increased
mortality, and reduced density of individuals
(this self-thinning effect is well established
for crop monocultures). Consequently, when equal
area samples are compared those in productive
habitats contain fewer individuals and thus by
chance alone, few species. Some support for this
from experimental productivity gradients that
show strong effects on plant density, and
simulation of random thinning to existing data
sets showing hump relationship.
12
Just how common is the hump-relationship?
Most recent review of diversity-productivity
relationship by Mittelbach et al (2001) looked at
171 published papers! For plants found that
predominant relationship between correlates of
productivity and species richness was a humped
(or unimodal) (40 of studies at all spatial
scales - from local lt20 km up to continental
scale gt4000 km) For aquatic systems, hump
relationship especially common Dodson et al
(2000) found evidence that phytoplankton,
macrophytes, and zooplantkon all showed unimodal
relationship. Not so for fish - diversity peaks
at high productivity - no observed decline. Why
would lakes commonly show hump relationships??
13
For animals, mostly no relationship of S with
productivity surrogates at local-landscape scale
(lt200 km) but increasing proportion of studies
found a humped relationship at larger scales (40
of studies at continental scales) Overall
Good support for a common occurrence of the
hump-shaped curve discussed in literature. Howeve
r, lots of variation (positive, negative and
U-shaped curves occur at all scales of
observation) - Many cases where no relationship
found between productivity and richness within a
habitat type, but a humped relationship when
studies included sampling across habitat types
(eg grassland and forest). Overall probability of
finding a hump increases as the scale of
variation in productivity increase (Mittelbach
2001).
14
Experimental approaches to investigating
diversity-productivity relationships.
Alter productivity (typically increased through
fertilization then examine diversity response).
Predict increased species density for low
productivity, decreased density for higher
productivity sites Are these experiments valid?
Processes of change in richness are related to
succession, colonization, adaptation, and
feedback linkages with soil microbes etc... All
long-term processes. Potential colonizing species
may be completely absent from widespread habitats
with similar productivity values
15
Not surprising therefore that fertilization
experiments have mixed results Lakes
Artificially manipulating productivity over the
range observed to produce a phytoplankton
diversity hump in natural systems failed to
produce the same relationship experimentally
(Dodson 2000) Land Most fertilization
experiments conducted in grasslands (recently
reviewed by Gough (2000)) - 11 of 16 studies
showed decline in species density in response to
N fertilization. 5 studies no response -
Magnitude of effect on species density was not
related to initial productivity (doesnt matter
where you are on the hump).
16
How about reciprocal effects? To what extent is
productivity itself dependent upon species
composition? Extreme example Coral reefs -
Rohde (1998) there is no reason to assume that
the waters in which coral reefs are found were
originally more productive than the surrounding
sea. Evolution of corals and calcareous algae
facilitated evolution of other primary producers
leading to an increase in overall
productivity If productivity is a consequence
of diversity then what will be the consequences
of loss of diversity?
17
Lot of research interest in ecosystem effects of
diversity came from the ECOTRON experiments
(Silwood Park, UK) - a community microcosm
including producers, consumers (2 levels),
decomposers
16 replicate chambers
http//www.cpb.bio.ic.ac.uk/ecotron/ecotron.html
18
ECOTRON experiments Effects of species richness
on ecosystem functioning. Effects of CO2 x
temperature interactions Microbial
diversity Naeem et al (1994) looked at ecosystem
processes including productivity correlates as a
function of biodiversity by establishing model
communities consisting of Primary producers (2,
5, or 16 spp) all self-pollinating, herbaceous
annual plants (weeds) Primary consumers (3, 4,
or 5 spp) - herbivorous insects and
snails Secondary consumers (1 or 2) insect
parasitoids Decomposers (3, 5 or 8 spp) -
earthworms and collembola Cant run all possible
spp combinations because of limited number of
chambers (this has been a major criticism)why??
19
Predators
Low and Medium diversity treatments are nested
subsets of the high diversity community
Herbivores
Plants
Decomposers
High
Medium
Low
Biodiversity
The key result - Lower diversity communities
were less productive and consumed significantly
less CO2, despite having equivalent population
sizes. - Variation in vegetation structure, or
plant community architecture, suggests a
mechanism for this effect.
20
- Higher diversity communities had greater
space-filling canopies and intercepted more
light, allowing higher photosynthetic rates and
productivity.
Low
Medium
High
21
Transmittance of light - a non-destructive
measure of productivity
Low
Medium
High
22
Recent studies have also looked at ecosystem
consequences of biodiversity in natural settings
Are more diverse sites more resistant to invasion?

• Kennedy et al. (2002) study with Tilman at Cedar
  Creek
• Show that for small grassland plots, the
  establishment (density) of invaders and the
  success (proportion of invasive plants that are
  large) are reduced.
• General question about this and ecotron
  experiment
• Is it diversity per se that generates these
  effects
• Or, is it species composition?

23
(No Transcript)
24
What is the relationship between species richness
and ecosystem function?
http//www.abdn.ac.uk/ecosystem/bioecofunc/intro
Symstad et al. (1998) Some evidence of
idiosyncracy? Looked at effect of species removal
on productivity at Cedar Creek - magnitude of
effect depended on species identity and order of
removal.