Spatial distribution of soil carbon and nitrogen storage and forest productivity in a watershed planted to Japanese cedar (Cryptomeria japonica D. Don)

Abstract

Digital terrain modeling was used to evaluate landscape-level spatial variation of soil C and N storage and site productivity in Japanese cedar (Cryptomeria japonica D. Don) stands. Soil C and N storage were measured in samples from surface soils (0–25cm depth) of 29 Japanese cedar stands in the 205-ha Myougodani watershed, Toyama Prefecture. The site index (C. japonica tree height at age 40 years) was used as a measure of forest productivity. Seven terrain attributes (elevation, slope gradient, aspect, profile curvature, plan curvature, openness, and wetness index) were calculated from a digital elevation model. Soil C and N storage were negatively correlated with slope gradient and positively correlated with openness. Variation in the site index was closely related to the wetness index. The prediction models using terrain attributes as explanatory variables explained 50% of the variability in soil C storage, 53% of the variability in soil N storage, and 75% of the variability in site index. This result demonstrated that this technique is useful for estimating the spatial distribution of soil properties and productivity in forest landscapes. On the other hand, there was no correlation between site index and soil C and N storage. Use of the prediction models in a geographic information system revealed that the spatial distribution of forest productivity differed considerably from those of soil C and N storage.