Measuring Population Size
A population is defined as a group of individuals of the same species, occupying a particular area at the same time. In all studies of quantitative ecology, it is essential to be able to estimate the number of organisms within a given area of ground or volume of water or air. In most cases, this is equivalent to estimating the population size; the methods employed are determined by the size and mode of life of organisms involved and also the size of the area under investigation.
For species living in uniform and accessible habitats, given their sizes are large enough, the population can be estimated by marking and recapture methods, or by using square grids of known areas. In some cases, when one is dealing with relatively small areas, it is possible to isolate the entire population and take a complete census. Although the methods of estimation are different, their outlying principles are simple and can be demonstrated in laboratory by using simple apparatuses.
Design Experiment I
Population size can be estimated by capture-recapture method.
Big bowl, beans, quick drying marker, small spoon.
- Filled the Petri dish with 500 beans.
- 52 rice grains were taken out by random and marked with the marker. They were allowed to dry.
- The marked beans were returned to the Petri dish and mixed thoroughly together with the unmarked beans. Here, all the beans represented individuals in a population, whereas the marked beans represented the first captured and then released individuals.
- A sampled of beans was removed by using a small spoon. The numbers of marked and unmarked beans were counted and recorded.
5. Step 4 was repeated 9 times.
The result obtained in the second attempt was invalid and therefore was discarded. This invalid result was mainly due to a small number of beans being removed in the recapture step. It suggested that a higher percentage of beans should be removed during the recapture step in order to get an accurate result. Furthermore, the percentage error could be calculated by the following procedure.
The total number of beans being used in this experiment was 500. Error = Actual no. – Calculated no. = 500 – 422 = 72 Percentage error = 78500 ? 100% = 15. 60%
Conclusion and Evaluation
The results obtains from this experiment suggested that capture-recapture methods could be used for estimating the size of population. This was because the principles behind this method were simple and no special technique or equipment was required. Limitations of Experimental Design and Suggestions for Improvement The model demonstrated in this investigation was much more simplified than actual field work.
More factors have to be considered when applying capture-recapture techniques in actual field studies. The following lists some of the factors:
- In actual field studies, sufficient time is needed between capture and recapture steps to allow random mixing. The less mobile the species, the longer the time lapse must be.
- The probability of capturing a marked animal may not be the same as that of capturing any member of the population.
- Marking may hinder the movement of the organisms or make them conspicuous to predators.
Suggestions for Improvement
When applying capture-recapture methods in field studies, many aspects need to be considered carefully and solved before accurate results can be obtained.
Design Experiment II
The number of organisms within a sample area (quadrat) can give an estimate of the total numbers in the whole area by simple multiplication.
Square grid model, macaroni, plastic bag.
- The model was placed on the floor to mark off 100 identical squares.
- The bag was filled with macaroni, which was then held 2 metres above the centre of the model.
- The bag of macaroni was emptied and scattered on the squares.
- 10 squares were selected randomly and the macaroni lying on these squares were collected separately.
- The macaroni in each square were counted and the recorded.
- The total number of macaroni released was counted.
- The operation was repeated two more times by selecting 20 and 30 squares respectively.
The average count, total number and the percentage error for each operation could be calculated by using the following formulae:
- Average count: mean number of macaroni collection: Sample calculation: 6+7+4+9+2+1+3+1+1+510 = 3. 9 per square
- Total number = Sample average per square ? 100 (all squares) Sample calculation of 1st operation: 3. 9 ? 100 = 390
- Error = Actual total – Calculated total Sample calculation: 331-390 = -59
- % Error = erroractual count 100% Sample calculation: 59331 100% = -17. 82%
Conclusion and Evaluation
The total population of a species in a whole area can be estimated first by investigating the distribution of this species in a small sample area and then multiplying the results by an appropriate scale factor. This was because the principles involved were simple and easily understood, but the results were accurate, with a low percentage error approximate to 5%.
Limitations of Experimental Design
It was clear that the percentage errors for the second and third trials were 1. 80 and 5. 38, which were within an acceptable range. However, the result of the first operation showed that it deviated from the actual total by nearly 20%, which was unacceptable. Further investigation suggested that the major cause for such a high percentage error was due to relatively small sample areas.
Design Experiment III
This method utilises the principle of diminishing returns in estimating population size.
If a series of successive samples are taken from a population, and the individuals are not returned, a decrease in numbers captured per unit effort is usually noted in the later samples. If the rate of decrease in captures per unit effort is constant, it can be measure and used to estimate the total population size.
The population size can be estimated by capture per unit effort.
Dry beans, plastic bowl, spoon.
- Filled the plastic bowl with 500 beans.
- The bowl was placed at shoulder level.
- A sweep of beans was removed from the bowl by using a spoon. The number of beans in the sample and the sample attempt were counted and recorded on a chart.
- The step was repeated until there were 3 unsuccessful sweeps.
- The above procedures were repeated two more times.
Conclusion and Evaluation
Based on the results from the experiment, the hypothesis that population size could be estimated through counting the number of individuals caught by successive captures.
Although the results showed that its accuracy was satisfactory with a percentage error of approximately 18%, this method had important practical uses in studying population of tiny organisms as they were difficult to observe or to be marked. Limitations of Experimental Design However, the method used in this experiment had room for improvement. The base of the container was hard, which definitely imposed difficulty in capturing the beans by a spoon, especially when there were only one or two layers of beans left in the container.
The undesirable effect was only very small number of beans could be taken out at a time at a later stage. This took a longer time to finish the experiment. Suggestions for Improvement If the floor of the container could be carpeted with a layer of soft material, for example a layer of sponge, the capturing of beans at the later stage would be easier because sponge is compressible. More beans could be taken out, meaning less time would be needed to complete the experiment. Similarly, a soft and durable spoon might help in the same way for saving time. Overall discussion on these census methods
Sample areas (quadrats), capture-recapture (mark-recapture) and capture (removal) are commonly used census techniques for estimating population size. Both methods are objective methods, involving direct observation only, without personal judgment. As indicated by the results of the previous experiments, the accuracy of these methods is similar. To determine which one is chosen for estimating population size will depend on the type of organisms being studied. Choosing suitable census methods The choice of the above mentioned methods for calculating population sizes depend primarily on the type of organisms being studied. Sample area techniques are frequently used in estimating population size of plants and sessile animals. These organisms will keep on staying in the habitat and investigators can have enough time to observe, to count and to record the data. However, this method is not suitable for fast-moving animals as they are always scared and will escape in the presence of an observer. Occasionally, this method is used in larger and mobile animals, such as deer, wild ponies and lions. In this situation, the size of the same area is not in square meters, but in square acres.
Capture-recapture techniques are commonly used in estimating population of fast-moving animals, or animals of secretive forms (e. g. bats, snakes). These kinds of animals will escape in the presence of an observer. Direct counts of number per unit area or unit volume are therefore impossible or impractical. Slow-moving animals are unsuitable for this method as they take a very long time for complete mixing during which deaths or births may occur. Furthermore, this methods applies successfully only to those populations belonging to the closed system. The numbers of individuals in a closed system remain constant over the study period.
On the other hand, capture-recapture methods cannot be used for the estimation of large populations because somewhere in the region of 20% of the population must eventually be captured and marked for accurate results. This usually needs a lot of manpower and money, making the method impossible. At the other extreme, animals with exceedingly low density populations, it is impossible to study using this method because insufficient numbers of individuals will be caught.
The capture (removal) method is very suitable for estimating numbers of small organism.
Their small size and mobility make it impractical to obtain exact counts of the number of individuals in unit area or unit volume sample. This method is particularly used in insect studies, because of their relatively shorter life cycle and their ability to shed the old cuticle during the marking process in the marking-recapture technique. Although each method may be used separately for estimating population size, it is preferable to use another one as a countercheck. Without the countercheck, nobody would know whether the estimation resulted from one method is understated, overstated or accurate.
Further discussion and comparison on capture-recapture, sample areas and normal techniques Capture-recapture techniques Capture and recapture methods can give a better estimation on animal population sizes compared to the other two methods. The technique and procedures involved in this method are comparatively more complicated and it requires capturing (usually by traps), adding marks (either colouring or plastic rings), releasing and then recapturing of the animals Special attention should be paid during marking and releasing steps. These are:
- The mark should not affect the animal The mark must be easily detected but not attract predators
- The mark must be durable
- The mark must not contain toxic preservatives
- Only healthy, unharmed individuals are released
- The elapse time between release and recapture to allow random mixing
- The marked individuals should not be overactive during the releasing period as this may attract predators.
Sample area techniques Sampling with quadrats is another widely applicable technique for obtained quantitative information about the structure and composition of terrestrial plant communities.
When doing quadrat analysis, the size of a quadrat should be large enough to contain significant numbers of individuals, but small enough so that the individuals present can be separated, counted and measured without confusion. Suggested quadrat sizes are one square meter for herbaceous vegetation, ten to twenty square meters for communities of shrubs and one hundred square meters for forest tree communities. Quadrat shapes are important in relation to the case of layout. In low vegetation, circular plots can be laid out very easily compared to quadrats which are rectangular and square shape.
With regard to efficiency, elongated, rectangular quadrats may give a more accurate analysis of the composition of a stand of vegetation rather than an equal number of square quadrats having the same area. The number of quadrats used in one study must, as a minimum, be sufficient to turn up the bulk of the species present in the stand. Investigations have shown that in a uniform habitat there comes a point beyond which analysing the species within a quadrat becomes unnecessary as it does not increase the number of different species recorded.
In other words, in a uniform habitat, there is a minimum quadrat number which can cover all the number of species recorded in that habitat. This suggests that sampling more quadrats beyond the minimum point is unrewarding and uneconomical on time. Finally, the quadrat method is subjective as it depends heavily on observation and personal judgement. This explains why estimates obtained by using quadrats are usually lower than those obtained from capture-recapture methods. For example, a single insect may be overlooked no matter how carefully the quadrats are observed.
With only nine insects seen, on insect overlooked is equivalent to a 10% underestimation. Capture techniques This method is suitable for estimating numbers of small organisms. Moreover, the population selected for this method must be small enough for sampling to remove a significant fraction of the total population. If the population is too large, little reduction in numbers capture per unit effort is likely. The curve (as shown in Fig 4), is nearly a horizontal line and so total population number is hard to obtain. Best results are usually obtained when the total number capture is at least one third of the total population.