EFFECT OF THE DESIGN AND THE SIZE OF INVE NTORY PLOTS ON THE ESTIMATION OF STAND CHARACTERISTICS

Effect of the design and the size of inventory plots on the estimation of stand characteristics. This paper aims at estimating the variance and bias of the estimation of characteristics of Pinus pinaster stands that are due to survey design and especially to plot size. Small surv ey plots (0.01-0.07 ha) similar to those used by the French National F orest Inventory (IFN) are simulated inside 39 large experimental plots , located in the Landes of Gascogne region in France. The stand charac teristics studied are: N (number of stems/ha), G (basal area), Cg (ave rage girth) and Co (dominant girth). The results show that small plots are not representative of the local stand conditions: the variance of N and G is high and much larger than that of Gg and Co; nevertheless, only Co is systematically biased (underestimated). When the stands ar e older and when the simulated plots are larger (0.05-0.10 ha), our re sults suggest that the larger plot size and the regularization of the structure of the stand induced by silvicultural practices diminish bot h the variance and the bias of the random error. These results indicat e that before using a growth model in order to project regional invent ory data if is necessary to check the compatibility between the data u sed to build the model and the data gathered in an operational survey.