Fig. 1

Examples of the decision tree-based bioclimatic models for xeric limits of the zonal forest stands dominated by a beech (basic predictors only), b beech (with EQ included) and c sessile oak. In the terminal nodes, bar diagrams visualize the probability of being above the aridity limit, with the partition of objects from the training data set among the leaf nodes on the top of the diagram boxes (n) See Table 1 for variable names