Analysis of plot-level volume increment models developed from machine learning methods applied to an uneven-aged mixed forest

We modeled 10-year net stand volume growth with four machine learning (ML) methods, i.e., artificial neural networks (ANN), support vector machines (SVM), random forests (RF), and nearest neighbor analysis (NN), linear regression analysis. Incorporating interactions of multiple variables, the ML methods ANN SVM predicted nonlinear system behavior unraveled complex relations greater accuracy than Investigating quantitative qualitative characteristics short-term forest dynamics is essential for testing whether desired goals in forest-ecosystem conservation restoration are achieved. Inventory data from Jojadeh section Farim Forest located uneven-aged, mixed Hyrcanian were used to model predict annual increment new technologies. The main objective this study was as preeminent factor yield models. In current study, two consecutive inventories 2003 2013 using techniques that physiographic input development: (i) (ii) (iii) (iv) (NN). Results various technologies compared against results produced ANNs SVMs a kernel function incorporated field-measurements terrain slope aspect variables able plot-level (94%) (87%). These provide compelling evidence added utility modeling context management.