Plant Ecological Methods

Sampling designs and Data Collection

• Choosing the right approach is a vital part of any

vegetation survey work.
 Plant Ecological Methods (Contd.)

Primary determinants of the sampling design:

Sampling designs differ in many respects. The most
important are:
1. Aims and objectives of the study;

2. Statistical properties: whether the sample plots can

be assumed independent of each other, and hence
whether formal statistical tests will be valid; i.e.,
proposed methods of data analysis and presentation
 Plant Ecological Methods (Contd.)

3. Type of habitat; (e.g., forestland, grassland, etc.)

4. Representation of dominant versus rare species, species

combinations, and combinations of environmental
conditions (and vegetation types);

5. Practical considerations: e.g. time needed in the field for

placement of sample plots; resource etc.
 Plant Ecological Methods (Contd.)

• Essential considerations pertaining to sampling;

– Quadrat; (square or circular)
– Sample plot;
– Area; (quadrat area)
– Sample number;
 Placement of sample plots
The following sampling designs will be discussed :
1) Selective sampling (cf. subjective or preferential)
• When the purpose of the study is description of selected
stands as in phytosociology;
• plots which are sampled selectively according to the
BrBl approach are called relevés.

• The representation of rare species and vegetation types

is often good.
 Placement of sample plots (Contd.)

• Selective sampling has not only been common within the

phytosociological tradition but has been the most common

sampling design for most kinds of ecological studies.

• The method is subjectively biased, (i.e., subjectivity will

constrain all subsequent analyses, and the demands of

independence of plots are not satisfied) but inclusion of

ecological measurements from each sample plot improves

environmental interpretations.
 Placement of sample plots (Contd.)

2) Stratified sampling
• Most types of vegetation studies, either deliberately
or implicitly employ stratified sampling.

• The principle of stratification is that the vegetation

of the area under study is divided up before
samples are chosen on the basis of major and
usually very obvious variations within it.
 Placement of sample plots (Contd.)

• Samples or quadrats are allocated to these different areas;

• Stratification is normally carried out by an initial

reconnaissance of the study area and/or examination of
any aerial photography which may be available;
 Placement of sample plots (Contd.)

• Major divisions are made primarily on the basis of

- Differences in growth form, (e.g., tree, shrub, herb);

- Physiognomy and structure of the vegetation (e.g.,

woodland, scrub, tall herb, grassland, open ground,
etc.);
 Placement of sample plots (Contd.)

• Other criteria for stratification,

- Different management regimes,

- Important environmental differences such as aspect,
geology or
- slope form or
- different degrees of successional change;
 Placement of sample plots (Contd.)

3) Random sampling
– Strict application of random sampling means that every
point within the survey area should have an equal chance
of being chosen on each sampling occasion.
How?
– Use of a grid of coordinates set up over the survey area
and pairs of random numbers taken to locate each
quadrat;
 Placement of sample plots (Contd.)

• From a statistical point of view the most optimal.

• No restrictions on statistical testing.

• Rare combinations of environmental factors or rare species are

poorly represented in the data set.

• This impacts the subsequent analysis, mostly in unwarranted way:

…%...
 Placement of sample plots (Contd.)

i) considerable redundancy for a few, dominant types,

while many rare but ecologically distinctive types
(species) are poorly represented or lacks
altogether;
 Placement of sample plots (Contd.)

ii) often results in a strong redundancy with respect to

mid-gradient conditions, while extremes are poorly

represented or not represented at all. (i.e. the rare

combinations of ecological factors are most often

associated with gradient extremes);

iii) much time is spent in collecting redundant information;

 Placement of sample plots (Contd.)

iv) unless the sampling intensity is very high, random

sampling often results in spatial clumping of plots
and an uneven coverage of the investigation area.
(this however mostly, not always, is very serious)
 Placement of sample plots (Contd.)

v) identification of sample plot positions in the field is often

claimed to be time-consuming.
 Placement of sample plots (Contd.)

• The disadvantages of random sampling mostly outweigh

the advantages.

• Most descriptive vegetation work, which is inductive

(reaching conclusion based on observation) in approach
does not require random sampling.
 Placement of sample plots (Contd.)

• Indeed much descriptive vegetation survey is biased

in its sampling and uses some form of stratified
sampling design.

e.g., Stratified random sampling

 Placement of sample plots (Contd.)

4) Systematic sampling
• Systematic sampling involves the location of
sampling points at regular or systematic intervals.
 Placement of sample plots (Contd.)

• The size of the sampling interval is extremely

important and usually is a fixed interval, such as
100 m or regular number of paces.
 Placement of sample plots (Contd.)

5) Transect Sampling

• The use of transect analysis involves systematic sampling

along a base-line in relation to a specific width and length

at predetermined distances along the line. (i.e., A line along

which samples of vegetation are taken)

 Placement of sample plots (Contd.)

• Usually set up deliberately across areas where there are rapid

changes in vegetation and marked environmental gradients;

• The transect approach is very popular in vegetation work.

 Placement of sample plots (Contd.)

Classic examples of laying out transects across gradients are:

• Up hillsides,
• Across major changes in geology,
• Through ecotones (such as terrestrial/aquatic/marine
transition zones); e.g. river and lake edges;
 Placement of sample plots (Contd.)

• For the determination of sampling units, the following

procedures have been specifically suggested for the
vegetation analysis on the respective terrain's indicated:
 Placement of sample plots (Contd.)

• The sampling plots will be placed along the line transect

at:
- 50 m drop in altitude for the sloppy terrain and

- 200 m intervals for the flat terrain.

 Placement of sample plots (Contd.)

a) Flat plains:

• A land terrain will be considered as 'flat' if it has a 0-10°slope

b) Sloppy terrain

• A land terrain will be considered as 'sloppy' if it has a slope

of greater than 10°
 Placement of sample plots (Contd.)

• The main purpose in using transects where the change in

vegetation is clearly directional is to describe the maximum
variation over the shortest distance in the minimum time.
 Placement of sample plots (Contd.)

• Transects are of two types:

- Line:
- Belt: where quadrats are laid out next to each other or
contiguously along a transect line;
 Placement of sample plots (Contd.)

6) Plotless
• In forest and woodland ecosystems and where
vegetation is sparsely distributed, as in semi-desert
or high alpine communities, the use of
conventional quadrat analysis and sampling may
be limited.
 Placement of sample plots (Contd.)

• In the case of woodlands, sampling of the tree

cover requires very large quadrats, and in sparse
communities, the large amount of bare ground
causes problems in recording.
 Placement of sample plots (Contd.)

• What is important here is the presence and absence of

individual plants and the distances between them.

• Several sampling techniques have been devised to

overcome these problems and are known collectively as
plotless sampling.

e.g. the nearest individual method,

 Vegetation survey and sampling

Vegetation description and measurement are derived from:

i. Morphological characters – i.e., structure

ii. Plant species present – i.e., floristic composition

The need for sampling:

Since the total area to be investigated can be too large to be

encompassed in one sampling,
- The actual vegetation description is based on a relatively
small piece of vegetation which is considered
representative of a larger unit.
Sampling is usually preceded by reconnaissance –

: an inspection of a local area where the vegetation

pattern with its dominant species and species
combinations is recognized and related to topography
and other apparent environmental conditions;
 2. Data Collection: sampling of species characteristics

Floristic data
• Each species within the sampling units will be recorded.

• Whenever possible, vernacular names of the species will

also be recorded;

• Plant species will be identified and checked at the

National Herbarium, Addis Ababa University. Voucher
specimens will also be kept at the National Herbarium
 2. Data Collection (contd.)

ii. Structural analysis

Information on the following structural attributes will be
collected.
• Diameter
– In each 20m X 20m plot all individual trees, bushes and
shrubs greater than 2.5 cm in dbh/dsh will be measured
for dbh with a diameter tape or tree caliper.
 2. Data Collection (contd.)

• Height
– In each 20m X 20m plot all individual trees, shrubs and
bushes will be measured for their height with a clinometer.

– Where slope, topography and/or crown structures make it

difficult to use height meter, heights will be estimated.
 2. Data Collection (contd.)

iii. Frequency
• cf. Frequency is defined as the probability or chance of
finding a species in a given sample area or quadrat.

• Recording frequency involves either throwing a series of

quadrats within a local area and recording the presence of
species in each (local frequency) or by using sub-units of a
quadrat and recording presence of a species in each sub-unit.
 2. Data Collection (contd.)

iv. Dominance
• Dominance for trees is conventionally measured as stem cover,
and stem cover is the same as basal area.
 2. Data Collection (contd.)

v. Diversity and Species Richness

– Species richness is taken as the count of the number of
plant species in a quadrat, area or community.

– Thus, species richness is derived from the floristic data

collected.
 2. Data Collection (contd.)

vi. Regeneration assessment

• All seedlings and saplings will be enumerated by species in
two 1m X 10 m sub-quadrats;
 2. Data Collection (contd.)

vii. Disturbance and human interference

– The extent of external pressure on the forest will be
determined following an arbitrary scale of disturbance of
the following activities
1) agricultural encroachment,
2) logging,
3) grazing,
4) charcoal production,
5) honey production
6) Fire
 2. Data Collection (contd.)

viii. Forest Site Description

• For each forest, the following parameters will be

determined:

1. Climate diagram

• A climate diagram from meteorological data from the

forest or the nearest station will be used to construct a

climate diagram for each forest.

 2. Data Collection (contd.)

• A climate diagram is a diagrammatic representation of the

average weather over the course of a year.

• A climate diagram is also a convenient tool that will enable

meaningful ecological comparisons between forests.

 2. Data Collection (contd.)

2. Geology and soil

• A geological description of the forest will be compiled from
existing sources or field surveys.

• The soils of the forest will be described in the same way as the
above.

• Physical and chemical properties of the soils will be determined to

enable a complete description of the environment for each forest.
 2. Data Collection (contd.)

• Data on soil aspects like

- soil depth,
- soil moisture,
- soil colour, and
- soil texture and structure will be collected as part of the
survey.
 2. Data Collection (contd.)

ix. Additional notes

• Other information on site conditions of study site includes
- altitudinal distribution of the forest vegetation,
- general aspect (exposure) of the forest,
- Slope, and
- information on accessibility of the forest from all
weather road.
Field Exercises
i) Select four species and assess them in terms of:
1. Density

2. Cover

3. Frequency

4. DBH

ii) The assessment is to be done along an altitudinal

gradient;
iii) Sample plot area – 400m²