Taxonomical

• Structure of an ecological community with respect to:

!Lecture 8_Community Ecology Lecture Slides, p.19
Magurran et al 2004 - "The bible on measuring diversity"

Richness-based metrics (such as species richness)

• Species richness is the most common richness metric
• Increases with sampling effort
  • Species accumulation curves: plot of number of samples vs number of species
  • only accurate if sampled at asymptote which can never really be reached
• highlights rare species - species with many rare species are over-emphasized
• Assumes entire species is sampled

$Chao$

• Useful when there are many undetected/invisible species
• This is an estimate of absolute number of species based on the rare species
  !Lecture 8_Community Ecology Lecture Slides, p.27
• Issue: If you have no double species, you need to use an "adaptation":

• $Chao2$ is another estimate that works with just presence/absence data

In-between metrics (between richness and evenness)

Shannon Index

• Assumes you have represented all species in the sample
  !Lecture 8_Community Ecology Lecture Slides, p.34
• Since it's a sum, you go over each species and sum them all up
• If you have a low number, that indicates a few species dominate the community
• Unconstrained, so it's not useful to compare two communities with different overall species richness
  • Ranges from $0$ to $ln(\text{community species richess})$
• Requires abundance data (does not work with presence/absense)

Evenness-based metrics

• (distribution of abundances), balance of species

In this example, they have the same richness, but Cat's Hill is more even:
!Lecture 8_Community Ecology Lecture Slides, p.20

Simpson's Index $D$:

• Probability that any two individuals drawn at random are from the same species
• This is a measure of dominance not evenness
• Maximum is 1, in a community with only one species, or close to that if you have a community completely dominated by one species
• Therefore, it's often expressed as $1-D$ or $1/D$ so that it increases with diversity
  !Lecture 8_Community Ecology Lecture Slides, p.37
• Or in R (with sample data from our example in slides):

v <- c(
  35, 26, 25, 21, 16, 11
)
D <- 0
for (i in v){
  D <- D + (i/sum(v))^2
}
> D
[1] 0.186233

Simpson's measure of evenness $E_{1/D}$

• Confusingly this is different from Simpson's Index
  !Lecture 8_Community Ecology Lecture Slides, p.41

Functional

• Measuring traits and how they function in the ecosystem
• Using a "functional space" plot (x = trait 1, y=trait 2), we can measure these three components:
• Three main components
  !Lecture 8_Community Ecology Lecture Slides, p.46

Functional Richness

• Total variation in the functional traits

Functional Evenness

• Decreases when:
  • Functional distances among species are less even
  • Abundances are less evenly distributed among species

Functional dispersion

• How far the abundance is from the mean of both

Phylogenetics

• Breadth of evolutionary history
• Essentially, you just add up the branch lengths on a phylogenetic tree