Development and validation of nonlinear height-DBH models for major coniferous tree species in Korea

This study was conducted to develop and validate the height-diameter at breast height (DBH) models for the three Pinus and one Larix species in South Korea using the six widely used nonlinear growth functions. The data collected from the 5th National Forest Inventory (NFI) of South Korea from the 2007 to 2008 measurement dataset were used in this study. A total of 11,693 total height and diameter measurements of the trees were used and split into two datasets. Of the dataset 80% was used for the initial parameters estimation of the six models, whereas the remaining 20% were applied to the model validation. On the other hand, the combined dataset (100%) was used for the final model fitting. The performance of the developed models was evaluated and ranked using the coefficient of determination (R²), root mean square error (RMSE), mean difference (MD) and absolute mean difference (AMD). The results showed that the R² of all six models for the four species ranged from 0.93 to 0.96. The lowest RMSE was 2.351 whereas the highest was 3.309. On the other hand, the overall MD ranged from -0.006 m to 0.044 m whereas the AMD ranged from 1.855 m to 2.626 m. Based on these four criteria, all six models were able to determine the height-DBH relationship and fitted the data well. The Modified Logistic and Korf/Lundqvist performed best compared with the other models using rank analysis. The equations developed for predicting the total height were used for the biomass estimation. With the exception of the Schnute model, all models provided similar results for all species, particularly for trees with less than 50 cm.