Influência do tamanho da parcela no desvio padrão num inventário florestal duma plantação de Eucalyptus grandis

Abstract

With a view to identifying an ideal plot size for carrying out forest inventories in plantations, an assessment was made of the influence of plot size on the standard deviation of a forest inventory in a Eucalyptus plantation. The study was carried out due to the need to define a parcel size that is representative to provide accurate, salistic volumes and that simultaneously its establishment leads to cost reduction, based on scientifically valid theories, the objectives of the inventory and local conditions . The study area was the Casa Lisa plantation, which is located in the Administrative post of Maluana, Manhiça district. The inventory consisted of establishing four different plot sizes. For each plot size, a sample of 25 plots distributed throughout the area was established using the simple random sampling system. In these plots DBH and tree height data were collected and used to determine plot volumes. The study consisted of calculating the standard deviation (SD) of the volumes obtained from the different plot sizes and with them the accuracy of the inventories was determined by the standard deviation and sampling error obtained. The larger plot, with 0.2 ha, offered the lowest standard deviation, which means that this size offers greater precision. In terms of plot sizes, it is observed that the increase in plot size causes a decrease in the standard deviation to tend to cancel out. The plot size at this standard deviation can be considered ideal, as increasing the plot area does not offer an increase in precision. Analyzing the coefficient of traffic (CV) it is verified that the stand is homogeneous, with a CV of 11% in the most precise plot and that the size of the trees follows the normal distribution. From this distribution, it is observed that in the plantation there are about 700 arv/ha, instead of the expected 1111 arv/ha. The average DAP of the stand is between 9-11 cm. The different sizes of plots produce volumes with differences and that only the plots of 0.05 ha and 0.2 ha produce volumes without differences. There is a strong dependence of the standard deviation on the plot size. This dependence is verified by linear and exponential regression analysis, which show that respectively 86% and 96% of the DP reduction is caused by the increase in plot size. The model's coefficient of determination shows that it better describes the relationship between the study variables. The linear model suggests that the ideal plot size is 0.23 ha. It is concluded that the size of the plot influences the DP of the inventory and that the larger plots offer greater precision than the smaller ones. Plots can be used, as they produce volumes without differences from the volumes obtained from larger plots. Therefore, the sampling intensity must be high in order to obtain greater representativeness and consequently greater precision (TRADUÇÃO NOSSA)