Development of ecoregion-based merchantable volume systems for Pinus brutia Ten. and Pinus nigra Arnold. in southern Turkey

Abstract

Estimating individual tree volume is one of the essential building blocks in forest growth and yield models. Ecologically based taper equations provide accurate volume predictions and allow classification by merchantable sizes, assisting in sustainable forest management. In the present study, ecoregion-based compatible volume systems for brutian pine and black pine in the three ecoregions of southern Turkey were developed. Several well-known taper functions were evaluated. A secondorder continuous-time autoregressive error structure was used to correct the inherent autocorrelation in the hierarchical data, allowing the model to be applied to irregularly spaced and unbalanced data. The compatible segmented model of Fang et al. (For Sci 46:1-12, 2000) best described the experimental data. It is therefore recommended for estimating diameter at a specific height, height to a specific diameter, merchantable volume, and total volume for the three ecoregions and two species analyzed. The nonlinear extra sum of squares method indicated differences in ecoregion and tree-specific taper functions. A different taper function should therefore be used for each pine species and ecoregion in southern Turkey. Using ecoregionspecific taper equations allows making more robust estimations and, therefore, will enhance the accuracy of diameter at different heights and volume predictions.

Simulation of over-bark tree bole diameters, through the RFr (Random Forest Regression) algorithm

“…According to the use of primary data from pine trees, the Max and Burkhart segmented taper equation under the nonlinear mixed-effects modeling technique was extensively used (TRICANDO and BURKHART, 2006;ÖZÇELIK and ALKAN, 2020). NICOLΕTTI et al (2020) evaluated the accuracy of bivariate and generalized linear mixed modeling in the representation of the Pinus taeda L. trunk taper, while many researchers tested parametric and semi-parametric models for constructing reliable taper models (ÖZÇELIK et al, 2016;ALGERA et al, 2019;MARCHI et al, 2020). As a step further, the scientifi c community that conducted research in this area, tested nonparametric modeling approaches, such as machine learning techniques and reported that it worth considering these methods as reliable alternatives to traditional taper-based equations (LEITE et al, 2011;NUNES and GÖRGENS, 2016;ÖZÇELIK et al, 2019).…”