Comparing resource-based and co-occurrence-based methods for estimating species niche breadth

    loading  Checking for direct PDF access through Ovid

Abstract

Questions

Examining the responses and co-occurrence patterns of 30 herbaceous forest plant species sampled altogether in 302 plots, we asked whether the niche estimates of rare and common species are affected by the different sample sizes for these species, and whether the results of the resource-based and co-occurrence-based methods coincide in the prediction of species specialization.

Location

Deciduous forests in northwest Germany.

Methods

We compare the classical method of determining niche breadth as the response of a species along environmental resource gradients (here soil pH and light availability) with a recently introduced method based on the co-occurrence of a target species with other species, using the Multi-Simpson index for turnover calculations.

Results

The total frequency of the species in the data set had a strong effect on the niche breadth estimates in the resource-based method, with species with higher frequencies having larger niches. In contrast, co-occurrence-based niche breadth was independent of sample size. When calculating the niche breadth estimates for equal numbers of plot occurrences for all species, none of the methods resulted in significant differences between common and rare species. Soil pH niche breadth and co-occurrence-based niche breadth were unrelated to each other, whereas light niche breadth and soil pH niche breadth, corrected for the different frequencies of pH values in the region, were positively correlated with the co-occurrence-based estimates. Species richness, increasing with increasing soil pH, was negatively correlated with niche breadth.

Conclusions

As the co-occurrence method is not distorted by varying plot frequencies, it offers a reliable alternative for extracting species niche breadth from vegetation surveys even if no environmental data are available. The gradient approach, however, remains important, as it provides additional information about measured niche characteristics (e.g. optima, response shape) along these gradients, which is of particular importance in various fields of applied ecology.

Related Topics

    loading  Loading Related Articles