PiscesLogoSmallerStill Searching for similarity

Top  Previous  Next

When we compare the flora or fauna sampled at different localities, we often wish to know how similar are their species assemblages. Numerous methods have been devised for the measurement of similarity, the most successful of which are described below. Legendre & Legendre (1998) give a more complete account of similarity and distance measures.


Similarity indices are simple measures of either the extent to which two habitats have variables in common (Q analysis), or variables have habitats in common (R analysis). Binary similarity coefficients use presence-absence data, and more complex quantitative coefficients can be used if you have data on species abundance. When comparing the variables at two localities, indices can be divided into those that take account of the absence of a variable from both communities (double zero methods), and those that do not. In most ecological applications it is unwise to use double-zero methods as they assign a high level of similarity to localities which both lack many species. We would not normally consider two sites highly similar because their only common feature was the joint lack of a group of species, which could occur because of sampling errors or because both sites were unsuitable.