Forma e tamanho de unidades amostrais para Sistema Agroflorestal em Monte Carmelo, Minas Gerais

Abstract

The forest inventory allows the collection of the desired information through sampling methods. The shape and dimensions of the sampling units are a decisive factor in a forest inventory. The objective of this study was to define the shape and size of fixed area sampling units most suitable for forest inventories of an agroforestry system in Monte Carmelo, Minas Gerais. First, the forest census was conducted, in which the diameter and height of trees were obtained. Subsequently, eight groups of fixed-area sampling units were distributed. Rectangular and circular shapes were evaluated, as well as areas from 100 to 700 m². For each sampling configuration, estimates of the number of trees (N), basal area (G), and volume (V), per hectare, about the census were evaluated. The sampling intensity was evaluated for each sampling configuration, taking into account an allowable error of 10%. The paired t-test was used to compare. Furthermore, the Kolmogorov-Smirnov (KS) test was applied to evaluate the statistical similarity of the diameter distribution between the census and each sampling configuration. The study concludes that areas of 300 m² are more efficient for estimating the number of trees (N) in both circular and rectangular shapes. For basal area, 300 m² was effective in circular plots and 600 m² in rectangular plots. In the case of volume, the rectangular area of 300 m² was more accurate, but in the circular shape, the 700 m² was more appropriate. Smaller areas require more samples for accuracy, while larger areas achieve the same accuracy with fewer samples. Rectangular units represented the diameter distribution well from 300 m², while circular units showed statistical differences in all areas.