You are here:
COMPARISON OF BIOLOGICAL COMMUNITIES: THE PROBLEM OF SAMPLE REPRESENTATIVENESS
Citation:
Cao, Y., D. D. Williams, AND D P. Larsen. COMPARISON OF BIOLOGICAL COMMUNITIES: THE PROBLEM OF SAMPLE REPRESENTATIVENESS. Presented at Ecological Society of America 2000 Annual Meeting, Snowbird, UT, August 5-10, 2000.
Description:
Obtaining an adequate, representative sample of biological communities or assemblages to make richness or compositional comparisons among sites is a continuing challenge. Traditionally, sample size is based on numbers of replicates or area collected or numbers of individuals enumerated and representativeness is based on the survey design as well as sample size. However, how well a sample actually represents the community from which it is drawn and what effect variation in the level of representativeness among sites has on site-to-site comparisons have been ignored. Sample representativeness (SR) literally means the similarity between a sample and its community in species composition and relative abundance. Without knowing the structure of a targeted community, the similarity among replicates randomly drawn from the community, i.e., autosimilarity, can be used to estimate SR. Using both field and simulated data we show that samples of equal size (area or number of individuals) from different sites differentially represent their communities. As a result, the proportion of total taxa richness at a site (%TTR) reached with a fixed sample size varied across sites and the site-to-site differences in TR changed with sample size. Consequently, a sample-size bias is introduced into the site-to-site comparisons. Using the same simulated and field data sets, we found the following: 1) Samples of equal size from different sites reached different levels of autosimilarity with lower levels reached in species-richer sites, indicating variation in how well samples of equal size represent their respective communities. 2) Autosimilarity positively and almost linearly correlated with %TTR indicating that autosimilarity may be a good predictor of richness. 3) Standardization of autosimilarity produced similar %TTR across different sites, i.e., the relative differences in taxa richness among sites becomes independent of sample size. Our preliminary results also indicate that standardizing on autosimilarity produced more stable comparisons of community composition among sites. We conclude that the standardization on autosimilarity can effectively reduce some of the undesirable effects of samples of equal size and improve the accuracy of the community comparisons.