Constructing single-entry stem volume models for four economically important tree species of Greece

Abstract

Three different nonlinear regression models were tested for their ability to predict stem volume for economically important native tree species in Greece. Τhe models were evaluated using adjusted R square (Adj Rsqr) root mean square error (RMSE) and Akaike information criterion (AICc), where necessary. In general, the quadratic polynomial and cubic polynomial models and the two-parameter power models fit the data well. Although the two-parameter power function fit best for fir, oak, and beech trees, the cubic polynomial model produced the best fit statistics for black pine. Making forest inventory estimates often involves predicting tree volumes from only the diameter at breast height (DBH) and merchantable height. This study covers important gaps in fast and cost-effective methods for calculating the volume of tree species at national level. However, the increasing need for reliable estimates of inventory components and volume changes requires more accurate volume estimation techniques. Especially when those estimates concern the national inventory, those models must be validated using an entire range of age/diameter and site classes of each species before their extended use across the country to promote the sustainable use of forest resources.

Keywords

• diameter at breast height
• fit statistics
• nonlinear regression
• tree species
• volume estimation