Sampling Techniques for Populations

A 2014 BBC report states that the world population of tigers is 3,000, down from 100,000 forty years ago. How are organizations, such as the London Zoological Society, able to determine this since animals do not stay still and allow themselves to be counted?

Ecologists, marine biologists and other scientists use numerous techniques to determine populations of wild animals and sea creatures. Some of the methods use analysis of past and current data, while others consider lifespans and migration patterns. There are also sampling techniques that are used to estimate populations, each suited to different situations. Two common techniques are segmenting an area and tagging.

Segmenting

Segmenting an area involves dividing a large area into smaller segments and then counting all the animals in the segment. One technique used to count the number of animals in a segment is to have a group of people walk through the segmented area and “herd” the animals as they walk. Observers count the animals as they leave the area. The number of animals in the entire population is extrapolated from the number of animals in the segment in a proportional manner.

Populations of small creatures, such as shrimp, can be estimated by scooping up a specific quantity of the creatures’ habitat, such a pond, and counting the animals in the scoop. Assuming the animals are evenly distributed over a specified area (an assumption that is often made but difficult to verify), the total population can be estimated based on proportional reasoning.

Tagging

In a standard tagging method, also called capture-recapture, a sample of animals from the population are captured and tagged (there are numerous ways to tag animals, including ear tags, leg bracelets and markings). These animals are then released back into their natural environment. At another time, another sample of these same animals is caught. The size of the population is estimated from the number of tagged animals that are caught again.

This method has been used in British Columbia, Canada, to estimate the population of grizzly bears. Tagging was done using DNA markers (British Columbia Grizzly Bear Population Estimate for 2012). With both of these methods, and other methods of estimating the population of animals, assumptions have to be made, including, as mentioned above, that the animals are evenly distributed over the area being studied. Using this assumption, a simple proportion can be used to estimate the population. For example, let’s say that in a hypothetical tagging (capture-recapture) study 100 animals from a certain area are caught, tagged, and released back into the area. Later, 50 animals are caught and checked to see if they are tagged; assume 20 are tagged.

Cross-multiplication can be used to solve for the unknown quantity:

20x = 100(50)
20x = 5000
x = 250

This means that approximately 250 of these animals are in the defined area. In practice, a combination of methods may be used to estimate a population of animals. But regardless of the methods used, we must remember that there may be unknown factors that could reduce the accuracy of the estimation. It is the expertise of the scientists doing the study that allows for better accuracy in the figures. An understanding of the various ways to estimate populations is important for these scientists so that they can use methods most suitable to the animals and the environment they are studying.

If you are interesting in learning more about sampling techniques, you may want to consider this career path and pursue a degree in data science.