To return to 1 — A, the simple definition of the index suggests relatively simple methods for its estimation by nonspecialists. One example is Cairns' linked estimator (CLE), a simple but accurate (unbiased) method for estimating the Simpson index by sequential comparisons (Figure 3). The method does not require organisms to be identified, or separated into groups. The first, and crucial, stage is to randomize the organisms in the sample. An individual is taken and compared with a second, recording only whether they look the same or not. The first is discarded and a third compared with the second, recording whether they look the same or not, the second is discarded and a fourth compared with the third, and so on. A 'run' is defined as a sequence of individuals which are considered to be the same, and ends when a new type of individual is encountered. The more runs there are for a given number of specimens, the greater the diversity. After considering N specimens, it can be shown that an unbiased estimate of 1 — A is
which, since only a series a simple comparisons are made, requires little, if any, taxonomic expertise. The error (imprecision) of the estimate depends on A and N, and for a particular Nis highest when A = 0.5, decreasing as it tends toward 0 or 1. Generally, for N < 100the error may be large, whereas for N > 200 the error tends to be small and the technique valuable. Although not widely used, this method has been proposed as a means to examine changes in diversity, in response to environmental perturbations, in groups which are speciose, numerous, responsive but difficult to identify to species, such as marine free-living nematodes.
Another example of the application of the Simpson index in a context which does not require the identification of species is the 'index of trophic diversity'. In essence, individuals are assigned to trophic groups (e.g., feeding types) rather than species, and the Simpson index is calculated. For example, an assemblage has a low trophic diversity when all individuals belong to the same feeding type, and a high trophic diversity when a number of feeding types are equally represented.
See also: Margalef's Index; Shannon-Wiener Index.
Was this article helpful?