Response of established oriental beech (Fagus orientalis L.) regeneration to canopy gap formation in the Hyrcanian forest ecosystems

Due to the lack of information about trees regeneration in the natural Hyrcanian forests, the current study focused on establishment of beech regenerations within natural canopy gaps (NCGs) in Galandrood forest. All gaps were classified into small, medium and large sizes. On the basis of collar diameter, regenerations were classified into three growth stages consisting of saplings, thickets and pole-sized regenerations. Stem number of each regeneration considered as density was counted in center and different sections of the gaps. Furthermore, the gaps border trees diversity indices were calculated, and the physiographic units were also recorded within the NCGs. Due to nature of the data, nonparametric tests such as Kruskal-Wallis, Eta and spearman's rank correlation were used for analyses in the current research. On the basis of the significant correlations of the regeneration's height to biological and environmental factors, a random forest (RF)-based predictive method was applied to simulate the regeneration's height. The results indicated that the pole-sized regenerations were not found in the small gaps and density was not significantly different among gaps; only height of saplings varied. Furthermore, the density of saplings and the height of pole-sized regenerations had significant variations among the sections of the geographical directions within the gaps. Sapling's height is significantly influenced by the physiographic units within the gaps. Significant correlations of gap border trees abundance to height of saplings and pole-sized trees showed that the regenerations within the gaps at early and late growth stages, unlike the middle growth stage, are probably more sensitive to shade conditions. On the basis of the RF simulation, the most important factors that influenced variations in the sapling's height within the NCGs were the border tress Shannon's diversity indices (H '), species dominance (D) and altitude. Moreover, Gap size and H ' had the highest relative effects on the thicket's height variations based on the results of the RF algorithm. Also, the results demonstrated that the abundance was the only influencing item for pole-sized individuals' establishment within the NCGs.