*2.4. Statistical Analysis*

We used cosine function convert aspect (ASP) and divided it into shade slope (<0) and sunny slope (>0). First of all, we have a Kolmogorov–Smirnov test for normality of the data. When the data satisfied the conditions of normal distribution, we used ANOVA and LSD to identify significant differences in vegetation and soil properties among the topographic conditions. We used regression analysis to establish the relative growth models to calculate plant biomass and used Pearson's correlation coefficient (*r*) to quantify the strength of the relationships among the vegetation, topographic, and soil indicators. All the statistical analysis was performed using version 3.5.1 of the R software (www.rproject.org).

We used redundancy analysis (RDA) to study the relationships between the vegetation and environmental indexes because the eigenvalue of the first axis of a detrended correspondence analysis (DCA) was less than 3 for the vegetation data. We performed these analyses using the "vegan" package for the R software (https://cran.r-project.org/web/packages/vegan/index.html).
