Species richness of vegetation types: fair comparison using standardization to sample completeness rather than sample area
| Authors | |
|---|---|
| Year of publication | 2015 |
| Type | Appeared in Conference without Proceedings |
| MU Faculty or unit | |
| Citation | |
| Description | For several decades, species richness has been one of the most popular diversity measures. However, its usage as a tool for comparison of richness between different communities might not always be appropriate. Traditional approach (i.e. comparison of equal-sized plots) may provide biased species richness values for communities differing by the level of undersampling. When comparing two communities differing by the size of species pool, a sample of given area may be large enough to fully characterize the less diverse community, but too small to sufficiently characterize the more diverse one. The problem of undersampling of such a community may be caused by limited number of individuals occurring in the area of interest, so a lot of species will simply not be captured by the sample of given size. As Chao & Jost (2012) suggested, instead of samples standardized by size, it is more reasonable to compare samples standardized by their coverage (a measure of sample completeness). We studied understory of six forest plant communities, with aim to compare diversity of herb layer using different ways of sample standardization: 1) by area (i.e. equally large vegetation plots), 2) by number of individuals, and 3) by sample coverage (a measure indicating the proportion of the total number of individuals in a community belonging to the species represented in the sample). With the intention of comparing usefulness of different standardizations, the six communities were selected with respect to differences in their alpha diversity, size of species pool and presumed density of plant individuals. For each of them, two types of data were recorded: a) vegetation plots of nested design (with subplots of 4 m2, 25 m2, 100 m2 and 400 m2) recorded at 12 randomly selected locations, and b) numbers of species and plant individuals in 100 0.04 m2 subsamples (placed evenly at the 100 m2 plot) recorded for 5 out of 12 plots mentioned above. For all three types of sample standardization, samples were rarefied to the same area, number of individuals and sample coverage, and species richness of each community was compared. Communities differ in their relative richness when using different types of sample standardization. Compared to less diverse communities, our results confirm our expectation that for fair comparison of species richness among vegetation types standardized by sample coverage, communities with higher species pool should be sampled by larger-sized plots. |
| Related projects: |