Diminishing influence of negative relationship between species richness and evenness on the modeling of grassland α-diversity metrics

Species richness and evenness have been widely used to investigate the spatiotemporal variation of alpha-diversity. However, some studies have indicated that a negative relationship exists between species richness and evenness. The question is how the differing sensitivity of alpha-diversity metrics and interactive behavior between richness and evenness affect the modeling of alpha-diversity variation. Here, we explored the response of species diversity, represented by three Hill numbers (i.e., species richness, exponential of Shannon index - expShannon, and inverse of Simpson index - invSimpson) focusing on the abundance of rare and common species, and Pielou index underlining the evenness of a community, to alpha-diversity variation through structural equation modeling (SEM). The model scheme integrated three categories of variables, spectral variation hypothesis (SVH), community pattern, and vertical structure, along the precipitation gradient spanning three steppes, including meadow steppe, typical steppe, and desert steppe. Our results showed that there were large differences in species richness across the three steppes, with v-shaped patterns emerging along the gradient (low-point in the typical steppe). Differences between steppes were diminished in the expShannon or invSimpson indices, though the v-shaped patterns persisted. The Pielou index showed the opposite pattern, with the peak in the typical steppe. Accordingly, a negative relationship between species richness and Pielou index was found across the three steppes. The concurrent increases in annual species number and dominant species abundance in response to precipitation variations led to the negative relationship. As a result, the SEM fitness on expShannon and invSimpson indices over the region was substantially diminished by the negative relationship. Overall, community pattern better explained the variation in species richness, invSimpson and Pielou indices. The performance of SVH differed among alpha-diversity metrics due to the collinearity with the variables of community pattern and vertical structure. This study emphasizes the variability of alpha-diversity metrics in response to environmental change. Particularly, distinguishing the asynchronous behaviors between species richness and evenness is paramount to account for alpha-diversity variation over heterogeneous ecosystems.