Concept of an Ecosystem

Lessons

Senior 6 · Biology

Concept of an Ecosystem

5 lessons

Lesson 2 of 5

Sampling ecosystems and ecological succession

25 min

How many plants are in a marsh?

The Rugezi wetland in the north of Rwanda covers thousands of hectares of waterlogged peat. No ecologist can count every sedge in it. Yet conservation decisions, such as whether a drained part of the marsh is recovering, need numbers. The answer is sampling: measure a small part of the ecosystem in a way that fairly represents the whole, then use statistics to judge how far the sample can be trusted.

This lesson covers the methods an ecologist uses, the calculations you will be asked to do in the national exam, and succession: the way a community changes, predictably, over time.

Definition

Sample: a small part of a population or area that is measured in order to estimate a feature of the whole.

Quadrat: a square frame of known area, often $0.25\,\mathrm{m^2}$ (50 cm by 50 cm) or $1\,\mathrm{m^2}$ , placed on the ground to sample plants or slow-moving animals.

Transect: a line laid across a habitat, along which samples are taken at regular intervals. A belt transect places quadrats continuously or at intervals along the line.

Species richness: the number of different species in an area. Species diversity takes into account both the number of species and how evenly the individuals are spread among them.

Choosing where to sample

	Random sampling	Systematic sampling	Stratified sampling
How the positions are chosen	Lay two tapes at right angles; use pairs of random numbers as coordinates, and place a quadrat at each	Place quadrats at regular intervals along a transect	Divide the area into zones and sample each in proportion to its size
Use it when	The area looks uniform, or you compare two areas; it avoids bias from choosing "interesting" spots	A factor changes across the area, such as soil water from a lake shore to dry land	The area has distinct zones, such as open water, sedge and grass in a marsh

What to record in a quadrat

Density: the number of individuals per unit area, $\text{density} = \frac{\text{mean number per quadrat}}{\text{area of one quadrat}}$ . Good for plants you can count one by one.
Frequency: the percentage of quadrats in which a species occurs. Quick, but a species with one plant per quadrat scores the same as one with fifty.
Percentage cover: the percentage of the quadrat's area covered by a species, seen from above. Used for grasses and sedges, where individuals cannot be told apart. A quadrat divided into 100 small squares, or a point frame of 10 pins, makes the estimate less subjective.

Animals need other tools: a sweep net for insects in grass, a pitfall trap for walking ground insects, a pooter to suck up small insects without harming them, a Tullgren funnel to drive small animals out of leaf litter, and kick sampling for invertebrates in a stream. Estimating the size of a whole animal population by capture and recapture comes in Unit 2.

Worked example 1: density, frequency and spread

A student placed ten random $0.25\,\mathrm{m^2}$ quadrats on a grazed slope and counted one herb species (illustrative data): 3, 5, 0, 2, 4, 6, 1, 3, 2, 4.

Calculate (a) the mean number per quadrat, (b) the density, (c) the frequency and (d) the standard deviation.

(a) $\bar{x} = \frac{\sum x}{n} = \frac{30}{10} = 3.0$ plants per quadrat.
(b) $\text{density} = \frac{3.0}{0.25\,\mathrm{m^2}} = 12$ plants per $\mathrm{m^2}$ .
(c) The species occurred in 9 of the 10 quadrats: $\text{frequency} = \frac{9}{10} \times 100 = 90\,\%$ .
(d) The deviations $x - \bar{x}$ are 0, 2, −3, −1, 1, 3, −2, 0, −1, 1. Their squares sum to $\sum (x - \bar{x})^2 = 30$ . So $s = \sqrt{\frac{\sum (x - \bar{x})^2}{n - 1}} = \sqrt{\frac{30}{9}} = 1.83 \text{ plants per quadrat (3 s.f.)}$

Interpretation: the standard deviation is large compared with the mean (about 61 % of it), so the plants are unevenly spread. More quadrats would give a more reliable mean.

Zonation along a transect

Where one abiotic factor changes steadily across the ground, the plants change with it in bands, or zones. Figure 1 shows a belt transect across the edge of a highland marsh. The results are drawn as a kite diagram: the width of each kite at each station is proportional to the percentage cover of that group of plants.

Figure 1: A belt transect and kite diagram across the edge of a marsh (illustrative data). The scale bar on the right shows the width of a kite at 100 % cover.

Read Figure 1 from left to right. Floating-leaved plants dominate the open water, sedges the waterlogged margin (peak cover 80 % at 20 m), grasses the damp ground (70 % at 50 m) and shrubs the driest ground (70 % at 70 m). The bottom row shows why: soil water falls from 95 % to 25 %. Each group lives where the soil water lies within its range of tolerance (Lesson 1) and where it out-competes its neighbours.

A correlation is not a cause. Soil water and plant cover change together along the transect, but so do soil oxygen, soil depth and height above the water. To show that soil water is the cause, you would need an experiment in which only soil water is changed.

Definition

Simpson's index of diversity: $D = 1 - \sum \left( \frac{n}{N} \right)^2$ , where $n$ is the number of individuals of one species and $N$ the total number of individuals of all species. $D$ runs from 0 (one species only) towards 1 (many species, evenly spread).

Spearman's rank correlation coefficient: $r_s = 1 - \frac{6 \sum d^2}{n(n^2 - 1)}$ , where $d$ is the difference between the two ranks of each pair and $n$ the number of pairs. It runs from −1 (perfect negative correlation) to +1 (perfect positive correlation).

Worked example 2: Simpson's index for two sites

Plants were counted in the same area of an undisturbed marsh margin (site A) and a drained, grazed part of it (site B) (illustrative data). Site A: five species with 40, 25, 15, 12 and 8 plants. Site B: three species with 85, 10 and 5 plants. Calculate $D$ for each site and compare.

Site A: $N = 100$ . $\sum \left( \frac{n}{N} \right)^2 = 0.40^2 + 0.25^2 + 0.15^2 + 0.12^2 + 0.08^2 = 0.1600 + 0.0625 + 0.0225 + 0.0144 + 0.0064 = 0.2658$ , so $D = 1 - 0.2658 = 0.734$ .
Site B: $N = 100$ . $\sum \left( \frac{n}{N} \right)^2 = 0.85^2 + 0.10^2 + 0.05^2 = 0.7225 + 0.0100 + 0.0025 = 0.7350$ , so $D = 1 - 0.7350 = 0.265$ .
Site A is far more diverse: it has more species, and its individuals are spread more evenly among them. Site B is dominated by one species (85 % of the plants), typical of a disturbed habitat. A community with high diversity is usually more stable, because the loss of one species affects the food web less.

Worked example 3: is sedge cover correlated with soil water?

Eight random quadrats on a marsh margin gave these results (illustrative data). Test the correlation at the 5 % significance level. The critical value of $r_{s}$ for $n = 8$ (two-tailed, $p = 0.05$ ) is 0.738.

Quadrat	1	2	3	4	5	6	7	8
Soil water / %	88	75	62	91	55	70	80	48
Sedge cover / %	70	52	48	60	30	55	64	38
Rank of soil water	2	4	6	1	7	5	3	8
Rank of sedge cover	1	5	6	3	8	4	2	7
$d$	1	−1	0	−2	−1	1	1	1
$d^{2}$	1	1	0	4	1	1	1	1

Null hypothesis: there is no correlation between soil water and sedge cover.
Rank each variable separately (largest = 1), find $d$ and $d^{2}$ (table): $\sum d^2 = 10$ .
$r_s = 1 - \frac{6 \times 10}{8(8^2 - 1)} = 1 - \frac{60}{504} = 0.881$
$0.881 > 0.738$ , so reject the null hypothesis: there is a significant positive correlation between soil water and sedge cover ( $p < 0.05$ ).

Condition: Spearman's test needs at least 5 pairs, preferably 10 or more, and shows correlation, not causation.

Warning

Safety and permission in wetland fieldwork

Work in groups, never alone, and stay on firm ground: peat and floating vegetation can give way under your weight. Your teacher decides where it is safe to go.
Wear boots and avoid skin contact with still or slow-moving fresh water, which can carry the larvae of Schistosoma (bilharzia). Cover cuts, and wash your hands with soap and water afterwards.
Protected areas such as national parks and protected wetlands need permission from the authority that manages them. Trample as little as possible, and put every stone and log back as you found it.

Succession: communities change over time

Leave a field uncultivated in Rwanda and within a few months weeds cover it; within a few years shrubs appear; given decades without fire, grazing or digging, trees would take over. This orderly change in a community over time is ecological succession.

Primary succession starts on bare ground with no soil: a new lava flow, bare rock after a landslide, or fresh sand. The first colonisers, the pioneer species, are lichens and mosses. They tolerate drought and bare rock, their acids and the freezing and heating of the rock break its surface into particles, and their dead remains add the first organic matter. A thin soil forms.
Secondary succession starts where a community has been removed but soil remains: an abandoned field, a forest after fire or clearing. It is much faster, because soil, seeds and roots are already there.

Each stage, called a seral stage, changes the conditions: deeper soil, more humus and minerals, more shade, more moisture. These changes let new species arrive that out-compete the ones that made the change. The succession ends in a climax community, stable as long as the climate stays the same; in most of Rwanda's highlands that would be forest. Figure 2 shows how biomass and diversity change.

Figure 2: Primary succession on a lava flow (illustrative). Biomass rises in an S-shaped curve; diversity peaks in the middle stages.

Diversity peaks before the climax because the middle stages hold a mixture of species from the earlier and later stages; at the climax, a few tall, dominant trees shade out many smaller plants.

A hydrosere is a succession that starts in open water. Sediment and dead plants slowly fill a shallow lake or pond: floating plants give way to sedges and reeds, then to marsh grasses, then to shrubs on the drier ground. The zones in Figure 1, from water to dry land, are a picture in space of that change in time.

Deflected succession. Where people or animals repeatedly interrupt succession, by grazing, burning, mowing or farming, the community is held at an earlier stage, called a plagioclimax. Pasture grazed by cattle stays grassland; on the savanna of Akagera, fire and browsing by large herbivores keep woody plants from closing the canopy.

A small fern growing out of a crack in a dark, rain-wet lava flow. Cracks trap water, dust and the remains of earlier pioneers, so they hold the first pockets of soil in which a plant can root. Photo: US National Park Service, public domain, via Wikimedia Commons

Bozeman Science: "Ecological Succession". Watch for the difference between primary and secondary succession, and note the pioneer species named at the start of each.

Tip

Exam technique for succession. Answer in a chain: "Pioneer species colonise → they change the abiotic conditions (soil depth, humus, water, shade) → the new conditions suit other species → these out-compete the earlier ones." Examiners reward the idea that each stage makes the habitat less suitable for itself and more suitable for the next.

✦

Practice: sampling and succession

State when you would use systematic sampling rather than random sampling, and give one example.
Explain why percentage cover, not density, is used to record sedges and grasses.
Twelve $1\,\mathrm{m^2}$ quadrats contained a total of 84 plants of one species, which was present in 9 of them. Calculate the density and the frequency.
Using Figure 1, state the distance at which sedges have their greatest cover, and suggest two abiotic factors, other than soil water, that change along this transect.
A site has four species with 25 individuals each. Calculate Simpson's index of diversity.
For ten pairs of values, $\sum d^2 = 36$ . Calculate $r_{s}$ and state whether the correlation is significant, given a critical value of 0.648.
Distinguish between primary and secondary succession, and explain why secondary succession is faster.
Evaluate the claim: "The climax community always has the highest species diversity."

Show answersHide answers

When an abiotic factor changes steadily across the area, so that you want to see how the organisms change with it. Example: a transect from a lake shore up to dry land.
Sedges and grasses grow in clumps and spread by underground stems, so single plants cannot be told apart and counted.
Mean $= \frac{84}{12} = 7$ per quadrat, so density $= 7\,\mathrm{plants\,m^{-2}}$ . Frequency $= \frac{9}{12} \times 100 = 75\,\%$ .
At 20 m (80 % cover). Any two of: soil oxygen, height above the water table, soil depth, soil organic matter, light (shade from taller plants).
Each species is $\frac{25}{100} = 0.25$ : $D = 1 - 4 \times 0.25^2 = 1 - 0.25 = 0.75$ .
$r_s = 1 - \frac{6 \times 36}{10(10^2 - 1)} = 1 - \frac{216}{990} = 0.782$ . $0.782 > 0.648$ , so there is a significant positive correlation.
Primary succession starts on bare rock or new ground without soil; secondary succession starts where soil remains after the community was removed. Secondary succession is faster because soil, nutrients, seeds and surviving roots are already present, so there is no need to wait for pioneers to make soil.
Not always. Diversity usually peaks in the middle seral stages, where species of early and late stages overlap; at the climax a few dominant trees can shade out many species, so diversity can fall a little. It is often, but not always, high at the climax.

Key points

Sample randomly for uniform areas, systematically (transects) along gradients, and stratified where there are zones; record density, frequency or percentage cover.
$D = 1 - \sum \left( \frac{n}{N} \right)^2$ measures diversity; $r_s = 1 - \frac{6 \sum d^2}{n(n^2 - 1)}$ tests correlation, which is not causation.
Succession: pioneers change conditions, later species out-compete them, and the community reaches a climax; biomass rises in an S-curve and diversity peaks in the middle stages.
Grazing, fire and farming deflect succession to a plagioclimax.

Sources and credits

8.2.7: Community Ecology, from Biology 2e by OpenStax (Mary Ann Clark, Matthew Douglas, Jung Choi), via LibreTexts Biology, licensed under CC BY 4.0. Used for ecological succession, pioneer species and the climax community. Adapted by Ganzaa.
Diversity index and Spearman's rank correlation coefficient, Wikipedia contributors, licensed under CC BY-SA 4.0. Used for the formulas. Adapted by Ganzaa.
Photo: 08Biocrust, SonoranDesertNPS, CC BY 2.0, via Wikimedia Commons.
Photo: Fern Growing in Lava Flow, US National Park Service, public domain, via Wikimedia Commons.
Video: Ecological Succession, Bozeman Science, embedded from YouTube.

Figures 1 and 2, the tables, the worked examples and the practice questions are written by Ganzaa. All data in them are illustrative. The critical value of $r_{s}$ for $n = 8$ was checked by exact calculation over all 40 320 rankings.

This lesson adapts CC BY-SA material and is shared under CC BY-SA 4.0.

1How many plants are in a marsh?

2Choosing where to sample

3What to record in a quadrat

4Zonation along a transect

5Succession: communities change over time

How many plants are in a marsh?

Choosing where to sample

What to record in a quadrat

Zonation along a transect

Succession: communities change over time