Comparing Methods for Developing Site Index Prediction Model of Sugi (Cryptomeria japonica) Planted Stand

Abstract

This study aimed to compare three methods for developing site index prediction model for specific species for a given site using environmental factors as explanatory variables. The following methods were compared in this study: (1) guide curve based method, (2) difference equation based method, and (3) combined site index prediction and height growth curve models method. The site index models were parameterized by the three methods with a permanent plot data and then evaluated for accuracy and practicability in model prediction using the stem analysis data. Root mean squared error and Spearman's rank correlation coefficient were used as statistics representing model accuracy and practicability; values obtained for the model derived from combined method were 4.468 and 0.754, respectively, which indicated that it was the best among the three models.