3.2. Relationship between environmental energy, climatic seasonality, water availability and richness pattern
Generalized linear models showed that the explanatory variables of environmental energy, particularly MAT and PET, were stronger predictors than that of climatic seasonality and water availability, and they showed a negative relationship to all richness groups (Table 1). MAT explained 6-19%, whereas PET explained 4-19% of the total variance in species richness. The proportion of variance also showed the significant role of environmental energy in determining species richness, particularly endemic-threatened (Fig. 2). In comparison with topographic and soil predictors, environmental energy, climatic seasonality, and water availability showed much lower contribution in the variation of species richness of all categories except endemic-threatened species richness, the environmental energy showed much higher contribution than soil type predictors (Fig. 2). Hierarchical partitioning modelling showed similar results to GLM models and suggested that environmental energy and climatic seasonality showed a high joined effect for all groups, while water availability had a highly independent effect (Fig. 3).