Modeling form factor of Fagus orientalis trees using nonlinear regressions in the Caspian forests of Guilan Province

One of the most important factors for determining the volume of trees is selection of an appropriate form factor. This study determined the most appropriate form factor formula and modeling the actual form factor in the Caspian region of northern Iran, Guilan’s forests. Totally, 150 beech (Fagus orientalis) trees were measured. Three coefficients of natural, artificial, and Hohenadel were compared with the actual form factor using pair-wise comparison test. Several nonlinear equations including logistics, Chapman-Richard, and exponential functions were used for modeling the form factor of beech trees. These functions were evaluated in terms of coefficient of determination (R2), paired t test, and accuracy statistics of ME, Root Mean Square Error (RMSE), MEr and RMSEr. The result showed that there were significant differences between the actual form factor and   the each mentioned coefficients. None of the models presented high R2 values. Accuracy statistics showed that exponential and logistic models can be usefully applied for estimation of the actual form factor coefficient. Also the results of t-test showed that there was no significant difference between of actual and estimated means of form factor derived from the logistic model. Therefore, logistic model was selected as the best for the estimating form factor of beech trees in Guilan’s forests of the Caspian region.