Assessing the diversity and composition of restored vegetation

Abstract

When surface-mined land is revegetated, the original vegetation should be restored in its full diversity. No quick, simple method exists for comparing the diversities of "old" and "new" plant communities. Current methods are laborious since they entail estimating diversity indices and their sampling variances. This paper gives a method for comparing the diversities of two communities directly, without estimating their diversity indices. It uses the fact that if one samples a community by listing the species at each of several sampling points, then, if community diversity is low, the several species lists will resemble one another closely, whereas if it is high, the lists will vary markedly among themselves. To compare diversities, one compares the pairwise resemblances between species lists within each community. If two communities are equally diverse, one wishes to know whether they have the same proportional species composition. If their diversities differ, one wishes to know whether the less diverse community is merely a subset of the more diverse. The data used for comparing diversities can be used to answer these questions also. Three examples illustrate application of the tests in comparing the vegetation on mined and restored land.