Austrian pine site indices as a basis for adaptive forest management in southwestern Serbia

Introduction The site index is the most commonly used indirect indicator of forest site productivity, based on the height-age relationship. The objectives of this study were: (1) to develop a dynamic site index model and (2) to create a site index map for Austrian pine (Pinus nigra) in southwestern Serbia using the Generalized Algebraic Difference Approach (GADA). Materials and methods The modeling was based on data obtained from stem analysis of 56 dominant trees (990 age-height pairs) and was validated using an independent set of 2,835 age-height pairs. In the initial stage, eight nonlinear models were tested, while the subsequent analysis employed the GADA approach to four selected models. Parameter estimation was performed using the Generalized Nonlinear Least Squares (GNLS) method with corrections for autocorrelation (ARMA (2,1)) and heteroscedasticity. The final model selection was based on a combination of statistical and biological criteria. Results The results showed that the Chapman-Richards, Hossfeld IV and Weibull models achieved a high goodness of fit ( R 2 adj ≈ 0.98), while the Gompertz model proved to be less suitable. However, only the Hossfeld IV model within the GADA approach fully satisfied the biological and theoretical assumptions, providing realistic asymptotic values and a biologically logical pattern of height growth. The constructed curves exhibited pronounced polymorphism, and the derived site index (SI) classes (SI 80 = 15–27 m) were consistent with the results of similar European studies. The generated site index maps revealed variability in site productivity. Discussion The developed site indices represent the first regional site indices for Austrian pine in Serbia and may provide a reliable basis for assessing site productivity and predicting height growth. In practice, the results may support the preparation of forest management plans and the planning of silvicultural measures adapted to differences in site productivity.