Statistical Analysis

The Czechanowski index (Gotelli and Entsminger 2001) was applied to quantify niche overlap between the species for all microhabitat parameters. The Czechanowski index ranges from 0 (no shared resource states) to 1 (identical habit utilisation). To test whether the observed niche overlap differed from a random pattern we carried out null model analyses in EcoSim 7.0 (Gotelli and Entsminger 2001). EcoSim simulates patterns of niche overlap and compares these randomised results with the observed data matrix. We used the algorithm RA3 (Winemiller and Pianka 1990) to test for non-random niche overlap. In this procedure, the observed niche breadth of each species is retained, but the utilisation of any resource state is allowed, including categories that were available but not used by the species. For each data set 10,000 replicates were created in the simulation.

To test for correlations between vegetation parameters, multivariate ordination analyses were performed. Because of the long gradient length in the data set, a correspondence analysis (CCA) was performed as implemented in the community ecology package vegan 1.13-1 for R (Oksanen et al. 2005). To evaluate the most important variables for the species, the locations of the species were fitted onto the CCA plot, using the function "envfit" with 1,000 random permutations. This method also generates an R2 measure and significance values based on the probability that random permutations would yield a higher degree of fit than the true data (Oksanen et al. 2005). The multivariate analyses were carried out in R 2.7.1 (R Development Core Team 2008).

Was this article helpful?

0 0

Post a comment