Estimation of tree height with machine learning techniques in coppice-originated pure sessile oak (Quercus petraea (Matt.) Liebl.) stands

In this study, in order to estimate total tree height, three different model structures with different input variables were produced through the use of 872 tree data points obtained from different development stages and sites in coppice-originated pure sessile oak (Quercus petraea [Matt.] Liebl.) stands. These models were fitted with machine learning techniques such as artificial neural networks (ANNs), decision trees, support vector machines, and random forests. In addition, the model based on DBH was fitted and its parameters were calculated using the ordinary nonlinear least squares method and this model was selected as the best model in Model 1. In other model structures, ANN model was chosen as the best estimation method based on the relative ranking method in which the goodness of fit statistics of the estimation methods were evaluated together. The inclusion of stand variables in addition to the DBH measurement in the model increased the R2 by about 36% and reduced the error rate by 55%.