Efeito da bordadura no crescimento diamétrico em povoamento de Eucalyptus urograndis, no estado de Alagoas

Abstract

Tree growth in forest stands can be influenced by spatial factors, especially edge position, which alters the availability of light, water, and nutrients, directly affecting diameter increment. This study aimed to evaluate the edge effect on diameter growth of Eucalyptus urograndis clones at 72 months of age, cultivated in an experimental area of 0.8 ha at the Engineering and Agricultural Sciences Campus of the Federal University of Alagoas, in Rio Largo – AL, Brazil. The stand was established at a 3 × 2 m spacing, containing three clones (TP361, I144, and VE41). Twelve linear plots were installed, each composed of seven trees distributed at different distances from the edge (0, 3, 6, 9, 12, 15, and 18 m). Trees were measured for diameter at breast height (DBH), obtained from circumference with a measuring tape, and for total height using a digital hypsometer. A total of 84 trees were analyzed using the SAS OnDemand for Academics software, with variable selection by the Stepwise method and linear regressions with different mathematical transformations. Three outliers identified by a Box-Plot graph were removed from the dataset. Analysis of variance (ANAVA) showed a significant edge effect on diameter growth (p = 0.0010), with trees located in the first planting row (0 m) presenting the largest DBH values, while those in the second row (3 m) were the most negatively affected due to competition from both the edge and the stand interior. From 9 m onwards, growth stabilized, demonstrating that the edge effect is restricted to the outer rows. The final adjusted regression equation presented an R2 of 0.9376 and a coefficient of variation of 26.25%, being considered suitable to estimate tree diameter as a function of edge distance. It is concluded that the edge effect must be considered in inventories, growth modeling, forest stand, and experiments planning, especially in small plots, where the proportion of trees affected by edges is more significant.