Testing of linear and nonlinear functions to fit the correlations
between variables was performed using Originlab 7.0 (Microcal
Inc., USA). Statistical analysis was performed using the Statistica
7.0 software package (StatSoft, USA). A one-way ANOVA followed
by a Tukey HSD test were used for comparison among horizons and
seasons. Differences among samples at P < 0.05 were regarded as
statistically significant. Analysis by general regression modelsmodule
of the above statistical software package was used to estimate
the proportion of variability explained by another variable and linear
fits with P < 0.05 were regarded as statistically significant. Mean
enzyme activities and microbial biomass content in soil samples
with specific moisture content were also predicted by linear regression
models. The data were derived from linear regression curves
and calculated as a difference between samples with moisture content
0.40 g g−1 versus 0.70 g g−1 in the L horizon and between 0.30
and 0.60 g g−1 in the H horizon and are always expressed as a percentage
of the value in the soil horizon material with the higher
moisture content. To test whether the fact that soil moisture effect
on individual variables was significant at P < 0.05, the 0.95 probability
limits for all fits were calculated and these fits were used to
predict the minimal significant effect of soil moisture on the studied
soil properties (Fig. 1). Surfer 7 (Golden Software, USA) was used