Stem volume models for Cryptomeria japonica of Jeju Island, Korea

Abstract

ABSTRACTThe accurate estimation of the stem volume of trees is an important task for forest managers. Two common variables used in stem volume models are the diameter at breast height (DBH) and the total height of a tree. These two variables can be used together, or only the DBH can be used. The present study was conducted to develop a simple volume model for Cryptomeria japonica found in Jeju Island, South Korea. The performance of six models using only the DBH and using both the DBH and total height as variables were evaluated. Fit statistics including coefficient of determination (R2), standard error of the estimate (SEE), mean bias (Ē), absolute mean difference (AMD), Akaike information criterion (AIC), and Akaike weight (AICw) were used to evaluate the models. Lack-of-fit statistics were also used for further evaluation. The results showed that the models using the DBH and total height as variables performed better than the models using only the DBH as a variable.