4.3. Data Analysis

The relationship between leaf stoichiometry of P. australis and soil properties was analyzed with a Spearman rank correlation. As our experimental set-ups were dealing with non-independent data or pseudoreplicates (multiple plots being the random factor, and different seasons being the fixed factor), and random effects of sample plots were tested using a linear mixed-effect model of statistical program R to determine sample independence [27]. One-way analysis of variance (ANOVA) was used to determine seasonal variations in soil properties. Significance analysis was performed using the Tukey post-hoc test. Redundancy analysis (RDA) was performed with a subset of environmental variables to assess the relative impact of abiotic factors on plant leaf stoichiometry, using explanatory environmental variables (SOC, STN, STP and their corresponding ratio, pH, SS, SWC, ST, AP, NO₃^--N and NH₄⁺-N) as regression covariates [2]. Detrended correspondence analyses of the datasets were undertaken before performing RDA to ensure that gradient lengths fit a linear model [3]. However, it should be noted that when there is no redundancy between the change of response matrix and the change of interpretation matrix, caution should be exercised. All data were logarithmically transformed before parameter testing and passed the homogeneity of variance test.