The effects of climate change on natural regeneration applications of oriental beech (Fagus orientalis Lipsky.) in the Devrek-Tefen district

In this study, which was carried out as a preliminary assessment in order to make some predictions and estimations about the effects of climate change in the natural regeneration area of Oriental Beech (Fagus orientalis Lipsky.) in three divisions (division of 70, 73 and 86) in Devrek-Tefen region, the mean of two-year-old natural beech juvenility was determined. Quantitative parameters related to regeneration success and adaptation ability such as mean height, mean root collar diameter and mean number of square meters were determined primarily by measurements and determinations. According to this, the mean height growth of 2 years old beech natural juvenilities is between 2.1-2.4 cm in all three divisions, the mean root collar diameter is between 1.1-1.3 mm, and the number of natural juvenilities in each three divisions in a mean square meter is changed between 10.5-10.9 items in meter square. According to the SSPs245 and SSPs585 scenarios, as a result of the calculations made by applying DeMortanne and Emberger indices, the pure oriental beech forests located in the semi-humid and humid forest belt in the Devrek region, which constitutes the research area, total 2156 ha in 2060 and 3448 ha in 2100, according to the SSPs245 global warming scenario. It is predicted that it will lose its natural regeneration conditions to a great extent due to the fact that it remains in the arid and semi-arid regions. In the research, widely used mechanical learning algorithms such as kNN (K-Nearest Neighbors), LVQ (Learing Vector Quantization), SVM (Support Vector Machine), RPART (Decision Trees) and RF (Random Forest) were used. In this direction, the mechanical learning algorithm, which predicts the effects on the natural youth of the 2-year-old beech in the 3 divisions in Devrek-Tefen, which constitutes the research area of global warming, with the highest accuracy level, with the accuracy levels of 74.3%, 75.61% and 73.56% calculated by the compartments. It has been RF. This algorithm was followed by kNN, RPART, SVM and LVQ, respectively, with their accuracy levels.