Lesson 6 of 10 · 20 min
Measuring species diversity
Calculate species richness, relative abundance, percentage cover, percentage frequency and Simpson's diversity index from field data, and use them to compare two sites.
What you will be able to do
- Determine species richness and the relative abundance (evenness) of each species from count data.
- Calculate percentage frequency and percentage cover from quadrat data.
- Calculate Simpson's diversity index, , and interpret its value.
- Compare the species diversity of two sites and judge whether the comparison is valid.
Two patches of open forest on the same hillside both have five kinds of understorey plant. Are they equally diverse? Not necessarily. In one, the five species might be present in similar numbers. In the other, a single grass might make up almost every plant. Ecologists use several measures, each answering a slightly different question.
Definition
Richness and evenness
- Species richness is the number of different species in a sample or area.
- Evenness (relative species abundance) is how similar the abundances of those species are. Each species' relative abundance is its share of all the individuals:
where is the number of that species and the total of all species.
Relative abundance of five understorey species (1 to 5) at two sites, A and B, with the same richness (100 plants counted at each). Site A is even; site B is dominated by species 1.
Both sites have a richness of 5, but site A is far more diverse: pick two plants at random there and they are probably different species. At site B they are probably both species 1.
Percentage frequency and percentage cover
Counting individuals is hard for plants that spread, such as grasses and groundcovers, where it is not clear where one plant ends. Two other measures avoid counting.
Key idea
Frequency and cover
Frequency says how widespread a species is; cover says how much ground it occupies. Cover is often estimated with a gridded quadrat: count the squares more than half covered by the species. Plants grow in layers, so the covers of all species in a quadrat can add to more than 100 %.
A quadrat divided into 25 squares. Shaded squares are more than half covered by a groundcover plant.
Use the gridded quadrat above. Calculate the percentage cover of the groundcover plant.
Worked example
Percentage frequency and cover from quadrat data
Eight quadrats were placed at random in a grassy woodland. The table gives the percentage cover of two species in each quadrat.
| Quadrat | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
|---|---|---|---|---|---|---|---|---|
| Kangaroo grass (Themeda triandra) | 35 | 0 | 20 | 45 | 0 | 10 | 25 | 5 |
| Blue flax-lily (Dianella caerulea) | 0 | 5 | 0 | 0 | 10 | 0 | 0 | 0 |
Calculate the percentage frequency and the mean percentage cover of each species, and compare them.
Step 1: Count the quadrats each species occurs in
Kangaroo grass: 6 of 8 quadrats. Blue flax-lily: 2 of 8.
Step 2: Calculate percentage frequency
Kangaroo grass: . Blue flax-lily: .
Step 3: Calculate mean percentage cover
Kangaroo grass: .
Blue flax-lily: .
Divide by all eight quadrats, including the zeros: the mean is for the whole site.
Step 4: Compare
Kangaroo grass is both widespread (75 % frequency) and covers more ground (17.5 % mean cover). Blue flax-lily is uncommon: found in a quarter of quadrats, covering under 2 % of the ground on average.
Try it yourself first, then check the working one step at a time.
Answer
Kangaroo grass: 75 % frequency, 17.5 % mean cover. Blue flax-lily: 25 % frequency, about 1.9 % mean cover.
Method
Frequency: quadrats containing the species ÷ total quadrats × 100.
Mean cover: sum of the covers ÷ total quadrats, zeros included.
Watch out for
Averaging cover over only the quadrats where the species occurs (23.3 % for kangaroo grass), which overstates it for the site.
Simpson's diversity index
Richness ignores evenness, and relative abundance gives a list of numbers rather than one. Simpson's diversity index combines both into a single value.
Key idea
What the index means
The fraction is the probability that two individuals picked at random (without putting the first back) are the same species. So is the probability that they are different species.
- : only one species, no diversity.
- close to 1: many species in similar numbers, high diversity.
- It rises with richness and with evenness.
Calculating SDI
Tabulate the counts
One row per species, with its count .
Find for each species
A species with contributes .
Add them to get , and add the counts to get
Divide by and subtract from 1
Calculate once, for the whole sample, not for each species.
Worked example
Compare the diversity of two open forest sites
Students counted ground-layer plants in ten quadrats at each of two sites in the same eucalypt open forest: a dry ridge and a moist gully.
| Species | Ridge, | Gully, |
|---|---|---|
| Kangaroo grass (Themeda triandra) | 42 | 6 |
| Spiny-headed mat-rush (Lomandra longifolia) | 18 | 31 |
| Blue flax-lily (Dianella caerulea) | 11 | 24 |
| Bracken (Pteridium esculentum) | 9 | 19 |
| Native violet (Viola hederacea) | 0 | 12 |
Calculate Simpson's diversity index for each site, to two decimal places, and compare the sites.
Step 1: Ridge: find and
. .
Step 2: Ridge: calculate SDI
Step 3: Gully: find and
. .
Step 4: Gully: calculate SDI
Step 5: Compare and explain
The gully is more diverse (). It has one more species (richness 5 against 4) and is more even: on the ridge kangaroo grass makes up of plants, while in the gully no species is more than .
SDI rises with both richness and evenness, so say which one (or both) explains the difference.
Step 6: Qualify the comparison
Both sites were sampled with the same quadrat size and number, so the comparison is fair. To judge whether is a real difference, repeat the survey (or calculate SDI for subsets of quadrats) and compare the spread of the values.
Try it yourself first, then check the working one step at a time.
Answer
Ridge ; gully . The gully is more diverse because it is richer (5 species against 4) and more even.
Method
For each site, find and .
.
Compare, and explain the difference with richness and evenness.
Check the comparison is fair (same methods, effort, timing).
Watch out for
Using or (the number of species).
Saying the ridge is more diverse because it has more of one species.
A pitfall trap survey in a patch of brigalow scrub caught these ground-dwelling invertebrates.
| Group | Ants | Beetles | Spiders | Crickets |
|---|---|---|---|---|
| Number caught | 10 | 6 | 3 | 1 |
Calculate Simpson's diversity index, to two decimal places.
Sample two sites and calculate richness and Simpson's index for each. Which changes the index more: adding a rare species or evening out the common ones?
Exam tip
Comparing two sites
The syllabus asks you to compare diversity in two spatially variant sites of the same classification: two patches of open forest, or two mangrove stands on different creeks. For a valid comparison:
- sample both sites with the same method, quadrat size, number of samples and counting criteria, at the same time of year;
- compare the same measure (richness, ) calculated the same way;
- judge whether the difference is bigger than the variation between repeat samples before calling it real;
- suggest causes from the abiotic and biotic data you collected (soil moisture, fire history, grazing), not from guesses.
Common mistake
SDI slips
- Using the number of species for . is the number of individuals.
- Dividing each by and forgetting to add, or adding the values and squaring.
- Forgetting the and reporting the fraction: a high value then means low diversity.
- Using and (another version of Simpson's index). Use the syllabus formula.
Summary
Species richness is the number of species; evenness is how similar their relative abundances () are.
Percentage frequency: the percentage of quadrats a species occurs in. Percentage cover: the percentage of the quadrat area it covers.
is the probability that two random individuals are different species; it runs from 0 towards 1.
SDI increases with both richness and evenness.
Compare sites only with the same methods, effort and timing, and check the difference against the variation between repeats.
Finished the lesson?
Try the 2 quick checks above and the lesson completes itself, or mark it complete now.