2 Plant surveying
Many of the problems that arise when surveying animals are caused by their mobility. Plants, of course, don’t present this particular difficulty, but they do present others. The two videos that follow show different sampling methods. As you watch, make notes on the techniques used in each one.
Transcript: Video 1
Transcript: Video 2
Counting corncockles is similar in principle to recording the numbers of animals (e.g. butterflies) present in a defined area. Surveying the chalk grassland and simply recording the plant species present is a useful way of recording the particular species present.
Activity 2 Assessing diversity
What are the limitations of recording the species present in assessing diversity?
Conspicuous species – maybe those with attractive flowers – are likely to be over-recorded and smaller, less conspicuous species will be frequently overlooked. Other species that might not be recorded are plants that have flowered earlier in the year and then died back. Some species may not yet have appeared and there may be species that are difficult to tell apart.
The data could be standardised (i.e. to help when comparing them with data from other sites) by using a fixed area to sample and a standard length of time in which to do the recording.
Activity 3 Standardising the sampling area
How was standardisation of the sampling area achieved in the surveys shown in the videos?
The area for sampling was defined by a square called a quadrat. A large quadrat was used in the case of the corncockles, where only one plant was being surveyed, but a much smaller one is needed if accurate sampling of every species present is required.
The quadrat used for the corncockle survey was 5 m x 5 m square. The survey provided an average number of plants in an area of 25 m2, which could then be used to estimate the number of corncockles present in the whole field.
In the corncockle survey it was possible to count directly the number of plants present, but this isn’t always possible so other measures of recording abundance are needed. Two methods that are often used are frequency of occurrence and percentage cover.
Frequency of occurrence is a measure of the chance of finding a particular species within a given sample area. For each square of a grid, the presence or absence of each species is recorded. The frequency of occurrence is the proportion of the total number of squares in which the species occurred.
Percentage cover, is a measure of the area of ground covered by a plant species and requires the use of a grid quadrat. The observer examines each square and makes a visual estimate of whether the square is fully occupied by the species or partly occupied. The partly occupied ones will be combined to give the estimated number of full squares they represent. Adding the number of fully occupied squares to this estimate gives the number of fully occupied squares out of the total number of squares in the grid. Other methods used to get percentage cover in plant surveys include a simple visual estimate of percentage by looking at the whole quadrat, rather than using a grid, and recording the percentage of each species in each sample. There are also various mechanical methods that can be used. For example, a bar with regularly spaced holes in it can be suspended above the area to be surveyed. A needle is put through each hole in turn. Each plant species touched by the needle as it goes through the vegetation is recorded as being present at this point. Photographs, too, can be useful tools in sampling.
Activity 4 Working out percentage cover
If a grid was five squares by five squares, nine of which were fully occupied by the target species, what would be the percentage cover for the species being surveyed?
A five-by-five square grid would contain 25 squares. If nine of these were fully occupied, the percentage cover would be nine as a percentage of 25, which is equivalent to 36 out of every 100, i.e. 36%.
So quadrats provide a way of formalising survey work. There is however another factor to consider, and that is the decision about where to place the quadrat. It would be very easy to bias your survey by not paying close attention to this criterion. For example, in Video 2 about chalk grasslands, throwing the quadrat on the ground appeared to provide a random approach. But actually there is still a bias at work here since there is a limit to how far you can throw a quadrat; only vegetation within reach is sampled.
Two sampling methods that aim to reduce bias are random and systematic sampling (Figure 8).
Random sampling is often used when it’s necessary to compare two or more sites. In this method, equal quadrilateral areas are marked out at each site and two sides of each area are numbered to give a coordinate system. The coordinates of quadrats for sampling are chosen at random. Then, by using a highly accurate differential GPS instrument (these are available for hire if you don’t have one), these randomly generated grid coordinates can be loaded into the machine and set out to centimetre-level accuracy at each site. One advantage of this method is that the same locations can be resurveyed over many years to look at changes over time.
With systematic sampling, instead of randomly placing quadrats, an area is surveyed by placing them at regular intervals, often linearly in an arrangement known as a transect. Transects are often used when surveying to see if there is a change in vegetation with distance.
Activity 5 Transect surveys
Can you think of any examples where transect surveys would be useful for surveying animal populations?
Transects are used to survey animal populations on sandy beaches, where a line of quadrats can be set out from the high-tide mark to the sea edge.
You will come across other examples of transect use as you read on.