Transcript Slide 1

Chapter 6: Construction
in The nature of the plant
community: a reductionist view
J.B. Wilson and A.D.Q. Agnew
Brooke Wheeler and Forbes Boyle
I. Theories: Models of Community
“Construction”
Deterministic—Plant interactions and environment control distribution of
species
OR
Stochastic—Communities are determined by random chance
Discreet—Communities are spatially seperated
OR
Continuous—There are no clear boundaries, community change is difficult to see
I. Theories
a) Deterministic and Discreet—
Clements’ view
(a)
(b)
b and c) Deterministic but
Continuous—
Whittaker’s view
(c)
b) Stochastic and Continuous—
Gleason’s view
b) Stochastic and Discrete
II. Frederick E. Clements
THE INTEGRATED COMMUNITY VIEW
“communities
were…nameable
and had ‘more or
less defining limits’”
“’An association is
similar throughout
its…general floristic
composition’”
“The same seral stage
may recur …with the
same dominants and
codominants.’”
II. Frederick E. Clements
THE INTEGRATED COMMUNITY VIEW
Wilson et al. (1996)
Objective: To test
hypothesis that the same
community will be found
in different locations.
Established “baselines” of neighbor quadrats to determine least similar pairs.
RESULTS:
83% of sites with “similar” vegetation occurred elsewhere…
So communities do occur.
III. Gleason
• Misunderstanding Gleason: his “views were identical
to Clements’.”
• Vegetation results from interference.
• Associations have limits and are separated by tension
zones (ecotones) (Gleason 1927) and communities
have boundaries and uniformity.
• Both Gleason and Clements thought of vegetation as
a mosaic of types.
• Gleason stated exact repetition of vegetation won’t
occur
IV. Whittaker and Austin
• Theory 2: continuous but deterministic
• Structure through competitive exclusion
• Co-evolution leading to avoidance
(dominants)
• Continuum Theory- vegetation organized
according to continuous change along
environmental gradients.
• JB Wilson et al 2004- flaws in analysis
cast doubt on continuum of change along
environmental gradients
V. Hubbell and Chance
• Functional (niche) equivalency
• Hubbell 2001- intended as a null model
not a best-fit model
• Section seems tossed in after it has
already been discussed at length
previously.
QUESTIONS
1) Were Clements’ and Gleason’s ideas about plant
communities develop essentially “identical” as
Bastow proposes?
2) Is “Clements vs. Gleason” simply a “straw man” for
paper intros?
3) There are 4 combinations of how communities
develop, but only 3 graphical representations.
What’s wrong with this picture???
QUESTIONS
4) What is the importance of gradient analysis?
5) Does it distract from the search for the nature of
plant community?
6) Does any statement in community ecology require
an appropriate null model?
VI. C-S-R Theory—Philip Grime
C (competition)
K
(ruderal) R
HABITAT
productivity
disturbance
r
S (Stress)
SPECIES
Highly competitive (C)
productivity
Stress-tolerant (S)
disturbance
Ruderal (R)
VI. C-S-R Theory--Stress
C (competition)
K
(ruderal) R
r
“The external constraints which limit
the rate of dry matter production of
all or part of the vegetation”
S (Stress)
Stress
Disturbance
VI. C-S-R Theory--Stress
According to Grime, STRESS is defined on
the COMMUNITY, not individual plants
Wilson and Agnew are critical of this:
--Tropical Tree dominants
Hubbell (2005): S species—shade tolerant
long life span
resistance to pests
Wilson argues: C or R species—rapid growth
post disturbance
VI. C-S-R Theory
C (competition)
DISTURBANCE:
K
(ruderal) R
r
“The mechanisms which limit the
plant biomass by causing partial or
total destruction”
S (Stress)
Stress
Disturbance
Little information given also
for competition
VI. C-S-R Theory and Succession
Grime: Productivity will influence successional trends
Agriophyllum squarrosum—a secondary pioneer
associated with dunes in semi-arid areas
Taraxacum officianale and Salsola kali—secondary
pioneers NOT restricted to desert communities
VI. C-S-R Theory and Succession
Wilson and Agnew: Difficult to separate out species in the R
and S corner of the Triangle, with regards to early succession
Only S species would be able to tolerate the S corner at all
seral stages of a community
RESULTS VARY!!!!
VI. C-S-R Theory Conclusions:
--”USEFUL GENERALIZATION”
--”DIFFICULT TO TEST”>>>NOT A WORKING MODEL
Questions:
1) Does overlaying MacArthur and Wilson’s r-K line help
explain the C-S-R triangle??
2) Why does Grime devote so little of this section to
disturbance and competion?
3) Is the species/character test section a useful addition in
this chapter?
VII. Tilman’s Theory
• “a hard center but woolly edges”
• Gaussian exclusion- coexistence of
competing species through different
limiting resources
• R* theory- in mixtures, species with the
lowest R* will outcompete others
• Deceptively simple
• Worked for alga in microcosm with
constant mixing (lab conditions)
R* theory- soil nutrients
• Soil: NPK major nutrients, vary in time and
space
• More difficult to support with experiments
• Tilman and Wedin (1991 a, b)- didn’t show that winner
could grow at lower N level and loser suffered in high
N as well
• Soil nutrients more complicated
• Decomposition, leaf leachate, precipitation
• Nutrients taken up by bacteria and micro-org.
• Animals redistribute nutrients
R* theory- soil nutrients cont.
• pH of soil affects processes and
availability of nutrients
• Varies with time- N more abundant in
spring
• Varies by nutrient
• P is immobile, plants must forage for it
• Localized depletion
R* theory- water, light
• Water
• Varies through time
• Varies with depth (rooting depth important)
• Light
• Vertical competition – decreases resource only for
individuals below
• Predicts that shade tol. species achieve tolerance by
lower light-compensation points but not supported
• Forest regeneration too complex for R*
VII. Succession
• Resource ratio theory of succession
• Successional position isn’t clearly related
RGR response to X10 nitrogen increase
1.35
Poa
pratensis
1.3
Schizachyrium
scoparium
1.25
1.2
`
1.15
Agrostis
scabra
1.1
0
2
4
6
8
10
N status in field: rank
Fig. 6.6: The experimental response to N compared to the rank of
species in a successional/N field gradient.
VII. Tilman conclusion
• These theories aren’t useful in the real
world
• While R* theory works well for microorganisms in labs, reality is complicated
• Tilman was brave to attempt to try to find
patterns in ecology
• R* conceived in a controlled,
homogenous, lab-tank environment.
QUESTIONS:
1) Are forests too complex for R* ?
2) Are there useful field applications of R* theory?
3) Is it impossible to see how to test or apply R*
outside of lab settings?
VIII. GRIME versus TILMAN??
or WILSON versus GRIME and TILMAN??
STRATEGY:
Plant Energy Expenditure fits
in nicely with both models
C-S-R—no species can
occupy all three points
R*--shoot versus root
strategy (ALLOCATE model)
SPECIES DIVERSITY:
Humped-back relation
between productivity and
species richness for BOTH
models.
VIII. GRIME versus TILMAN??
or WILSON versus GRIME and TILMAN??
COMPETITION:
Grime>>>Low in Stressed Environments
Tilman>>>Equal Across Stressed and Unstressed Environments
Wilson>>Interference limits plant abundance across all environments.
--No competition in early succession (Clements)
--Spatial mass effect
--Herbivory limits abundance
VIII. GRIME versus TILMAN??
or WILSON versus GRIME and TILMAN??
Wilson:
Both Grime and Tilman models do not account for resource
versus non-resource factors in their environmental gradients.
Difficult to test hypothes between Grime and Tilman models
because of the “GROWTH RATE ARTIFACT”:
--vegetation planted in pots will come to competion
earlier in high productive environments (Grimes).
VIII. THE WILSON and AGNEW
HYPOTHESIS!!!
Beta Niche Gradient (non-resources): competition is constant
Alpha Niche Gradient (resources): competion is strongest when
resource is in shortest supply
In High Stress Environments (deserts):
Very similar to Tilman’s idea that competition intensity is constant
across ALL communities
VIII. THE WILSON and AGNEW TEST!!!
Test the degree of competition
(Relative Growth Rate) along
the S-C gradient
C (competition)
K
(ruderal) R
r
S (Stress)
CONCLUSIONS:
1) Competition is equally intense along a non-resource gradient
2) Severest competition occurs at low-levels of a “sought after” resource
QUESTIONS:
1) How does Clements’ model of community development
differ from Grime’s and Tilman’s?
2) Why does Wilson come to support Clements’ model and
refute Grime’s and Tilman? Is a reason made clear in the
chapter?
IX. Synthesis
1. Too soon to tell
•
•
Community ecologists are in the worst
position
Dismal depiction of how little we know
2. Does vegetation suit our models?
•
•
We love Clements (in case you had not
heard)
Variation along gradients is continuous or
discontinuous due to a switch
IX. Synthesis cont.
• Grime’s C-S-R theory- useful
generalisation
• Tilman’s R* theory too simplistic
• “Does the vegetation suit our models?”
approach to plant ecology
• Complexity makes it difficult for vegetation
to fit simple models
Box 6.1: Types of interaction between plants.
At the species (or within-species) level
negative effects
interference (negative effects via reaction)
competition: species X removes resources from the environment, which
are then unavailable to species Y
allelopathy: X produces a substance toxic to Y
spectral interference: X changes the red/far-red balance, disadvantaging Y
switch: X causes reaction in an environmental factor, disadvantaging Y
negative litter effects: X produces litter of a type that disadvantages Y
(positive effects are a type of subvention)
parasitism: X removes resources directly from Y
autogenic disturbance: X disturbs, disadvantaging Y
negative effects via heterotrophs: X changes the heterotroph population, disadvantaging Y
Subvention (positive effects)
mutualism = X and Y both benefit relative to their being at the same density on their own
benefaction = X benefits Y as above, with no known advantage/disadvantage to itself
facilitation = X benefits Y, to its disadvantage
At the community level
guild/community X gives a relative disadvantage to itself:
the effect is density-independent: facilitation and/or autointerference = relay floristics
the effect disappears at low density of X (negative feedback) = stability
guild/community X gives a relative advantage to itself = switch
Three Things
1. Plant communities generally have many
species.
2. Heterogeneity is a rule.
3. In order for there to be “science in plant
community science” we hope there are
“rules governing the assembly of species
in them”.
IX. Heterogeneity
• Heterogeneity of environment- future
community process research should
concentrate where allogenic heterogeneity
is low
• Call for work on plant/littler effects on soil
• Ignore species area curves because they
don’t tell us much
IX. Assembly rules
• Evidence mostly from herbaceous
communities- esp. Otago Botany Lawn
• Difficult to search for assembly rules
– Don’t know what the rules are
– Need character-based rules (careful
selection)
• Preadaption of species is key
IX. Conclusions
• Need for integrated knowledge of plantplant interactions
• Switch is supreme process in plant
communities
– Move beyond the “easy task”
– Switching causing Alternative stable states
Questions
• Do plant ecologists just produce models
and try to make the data fit?
• Do species area curves provide useful
insights into community ecology?