Evaluation of species richness estimators in studies of diversity involving two larval digenean communities parasitizing snail hosts

Abstract

The nonparametric estimators of species richness are some of the most widely used extrapolation methods in studies of biodiversity. These studies need relative large samples to achieve total diversity (including rare species), and estimators are routinely used to alleviate this problem. In the existing literature, parasite communities have received little attention. In fact, the present study is the first dealing with parasites in invertebrate hosts. Real data on seasonal diversity of larval digeneans in the snails Heleobia conexa and Heleobia australis from Mar Chiquita coastal lagoon (Argentina) were used to evaluate seven nonparametric estimators of species richness, observing the behavior of each method and its ability to estimate the richness using different subsample sizes. In addition, estimated species richness was compared with the previously observed species richness in the parasite component communities of both snail hosts, and the minimum sample size, necessary to include all the observed species of larval digeneans, was estimated using the nonparametric estimators. The bootstrap method had the best overall performance in the two communities. Although this estimator required large subsamples to estimate sample richness, it was in general the least biased and most accurate, and it presented the most similar curve shapes with lack of erratic behavior through the seasons. The results obtained from the comparison between the minimum sample sizes estimated in previous studies of both snail hosts and those suggested by the estimators also support the use of species richness estimators as methods to calculate the minimum sample size in studies of diversity.