Abstract
Selective sample plots of 0.01∼1.0 ha and square, circle, and diamond in shape were placed at random in 4 ha model stands consisting of survey data from actual Japanese cedar plantations. From this computer simulation, we developed an estimation model to determine the coefficient of the stand volume variation in the sample plot based on the plot size and shape and the stand density. Assuming aiming precision of 10%, the recommended plot size is 0.1 ha for a young stand of about 2,000 stem/ha, 0.2∼0.3 ha for an immature or mature stand, and 0.4 ha or larger for a final aged stand of about 500 stem/ha, respectively. If more than 270 trees are included in the sample plot, the precision declines below 10%, regardless of plot shape and stand density. Particularly in the case of a small plot, the coefficient of variation by square plots exceeded that with circle or diamond varieties, when the sample plots were placed at random, regardless of stand disposition. More accurate results can be expected from the circle or diamond plot than the square plot, regardless of the skill of the surveyor selecting the sample plot.
