Geostatistical Analysis of Soil Water Content in Natural and Secondary Forests of Changbai Mountain, China

Abstract

Applying a combination of classical and geostatistical analysis, we investigated the distribution properties of soil water contents in broad-leaved Korean Pine forest (A) and secondary Polar-Birch forest (B) in Changbai Mountain, Northeast China. The results showed soil water content was significantly different between plots A and B (P < 0.05). The variation coefficient (CV) for soil water content was lower in plot A than in plot B at 0-10 cm depth, while the CV was higher in plot A than in plot B at 10-20 cm depth. Geostatistical analysis revealed that the exponential models were the optimality models for the soil water content in both plots. The spatial structured variance accounted for the largest proportion of total variance in soil water content in both plots. The autocorrelation range value for the semivariogram of soil water content was 8.9 m at 0-10 cm depth and 12.6 m at 10-20 cm depth in plot A, which was more than those in plot B (8.0 m) at both depths. Maps obtained by kriging showed that soil water contents in both plots exhibited different spatial distribution patterns. It could be conclude that the differences of rain time, precipitation form, topography and forest type may contributed to the distribution of soil water content in Changbai Mountain.