Transcript Competition between two Arabidopsis genotypes

Measuring plant abundance
and ecological processes
with the pin-point method
Christian Damgaard
Department of Bioscience
Aarhus University
Bioscience – Aarhus University
Plant abundance data
Next generation ecoinformatics
From plant occurrence to plant abundance and growth
Predictive process-based ecological models
Refining classic ecological hypotheses
Need to develop the measurement framework
Special data structures
True statistical inferences
Bioscience – Aarhus University
The pin-point (point-intercept) method
Method for measuring:
i) cover
ii) vertical density
Place a frame with a grid pattern
A pin is inserted vertically through one of
the grid points into the vegetation
The pin will typically touch a number of
plants and the different species are
recorded (to determine cover)
The number of times the pin hits the same
species is also recorded (to determine
vertical density)
This procedure is repeated for each grid
point
Bioscience – Aarhus University
Hierarchical plant cover data
Plant cover data has too many zero values and too
much variance compared to a binomial distribution
U- shaped distributions of plant cover are typical
Large-scale ecological processes (among-sites):
environmental drivers
extinction / colonization of sites
Small-scale ecological process (within-sites):
size of individuals
density-dependent population growth
inter-specific competition
Bioscience – Aarhus University
Beta-binomial distribution
q : mean cover at the site
 : intra-plot correlation
𝑓 𝑦 ~ 𝐵𝑖𝑛 𝑛, 𝑣 ;
𝑛
= 𝑦
1
𝜈 ~ 𝐵𝑒𝑡𝑎
1−𝑞 1−𝛿
𝜑 𝑞 𝛿−1 ,𝑦 𝜑
𝛿
1
𝜑 𝛿−1,𝑛
𝑞
𝛿
− 𝑞,
,𝑛−𝑦
 ( x, n)  ( x  n) ( x)  x( x  1)( x  n 1)
E (Y )  n q
Var (Y )  n(1  q )q (1   (1  n))
Bioscience – Aarhus University
1−𝑞 1−𝛿
𝛿
Relationship between plant cover and
intra-plot correlation in dune grasslands
Bioscience – Aarhus University
Important to include spatial correlation
The cover of Calluna vulgaris and
Deschampsia flexuosa on dry heathlands
Bioscience – Aarhus University
Important to include spatial correlation
If  was set to zero (no spatial correlation) then there
was a strong significant effect (grey line, P < 0.0001)
If  was allowed to vary then the effect was found to be
insignificant (black line,  = 0.57, P = 0.19)
Bioscience – Aarhus University
Separation of process and sampling variance
Process equation (solid error)
Dxi =a + ei
ei ~ Normal(0,sp2)
Measurement equation (dashed error)
yi,j ~ Beta-binomial Distribution (n, xi, )
Bioscience – Aarhus University
Erica tetralix on wet heathlands
a
Bioscience – Aarhus University
sp2
Is the cover of Erica tetralix regulated by
nitrogen deposition or pH or both?
Bioscience – Aarhus University
Conclusions
 The use of relevant parametric distributions are needed to
test ecological hypotheses
o Erroneous conclusions, if spatial variation is ignored
 State-space models and SEM are flexible tools for
modeling different ecological processes
o The measurement equations can be tailored to
accommodate different measures of plant abundance




pin-point method
Visual estimation, Braun-Blanquet, Hult-Sernander
Böcher-modified Raunkjær method
…
 Second-order statistics – from pattern to process
o Two species
o n species
Bioscience – Aarhus University
Evidence for competitive interactions in
plant communities
How often has the competitive interactions in a plant
community been measured in a natural plant community
compared to the times competition has been postulated
to be an important ecological mechanisms?
1/1000 is certainly too high
1/1000,000?
Bioscience – Aarhus University
Measuring ecological success
Natural and semi-natural plant
communities are dominated by
spatially structured perennial
species with variable life histories
Often difficult to count individual plants
Large size variation among individuals of the same species.
The ecological success may be assessed using the pin-point method
cover (= relative area that the species cover)
vertical density (= 3D space occupancy ~ biomass, plant volume, and LAI)
Bioscience – Aarhus University
Plant cover and vertical density – assumptions
i) Plant cover and vertical density measure ecological success
ii) Due to the growth form of most plant species, the vertical
density will increase relatively faster than plant cover during
the growing season
iii) Species with a high cover in spring will have relatively high
vertical density at the end of the growing season, however, the
vertical density is reduced by the cover of other species due
to competition
iv) A plant species that grows to a
relatively high vertical density has a
relatively high cover the following year,
i.e., plants allocate resources into
occupying resource space the following
year
year t
Bioscience – Aarhus University
year t 1
Plant cover and vertical density – model
Bioscience – Aarhus University
Case study: dry heathlands
Ambient treatments in the CLIMAITE experiment
Calluna vulgaris and Deschampsia flexuosa are the
dominating species and are expected to compete for
resources
Bioscience – Aarhus University
Significant competitive interactions between
C. vulgaris and D. flexuosa
Bayesian posterior distribution of the parameter that measures the
competitive effect of C. vulgaris on D. flexuosa within growing
seasons
MCMC – 100.000 iterations
Bioscience – Aarhus University
Predicted change in plant cover
Inserting samples from the joint posterior distribution into the
model makes it possible to test compound hypotheses
Assuming that the measured competitive interactions were
unaltered C. vulgaris was predicted to outcompete D. flexuosa
However, individuals of
C. vulgaris becomes
senescent after about
30 years
Bioscience – Aarhus University
“Demography” of space occupancy
For a selected species there are four possible
successive pin-point recordings at a fixed position:
1. the selected species was hit both years
2. the species was not hit in any of the years
3. the species was only hit the first year
4. the species was only hit the second year
Event #3 – species has disappeared from the space –
possibly due to death?
Event #4 – species has appeared at the space – possibly
due to colonisation or growth?
Bioscience – Aarhus University
Quantifying the importance of survival and
colonisation
The maximum likelihood estimates of the probabilities
of “survival” and “colonisation” are calculated using
the expected probabilities of the four possible events
(Damgaard et al. 2011)
The elasticity of the change in plant cover to both
survival and colonisation are calculated, i.e., is the
change in cover primarily caused by altered mortality
or altered recruitment?
Bioscience – Aarhus University
Conclusions
 Longitudinal species abundance data are needed for
measuring ecological processes, e.g. growth,
competition, mortality, and space colonisation
 Joint posterior distributions of the parameters in a
population model allows the examination of compound
ecological question, e.g. probability of extinction
Bioscience – Aarhus University
In collaboration with:
Inger Kappel Schmidt
Johannes Ransijn
Beate Strandberg
Knud Erik Nielsen
Morten Strandberg
and others…
Bioscience – Aarhus University