Comparing the methods for quantifying the culm surface area of bamboo, Phyllostachys pubescens

We compared several methods for quantifying the culm surface area (S) of one of the most common bamboos in Japan, Phyllostachys pubescens Mazel ex Houz. Nine sample culms of P. pubescens were felled, and the true S was determined by the fine resolution analysis of the culm form (S FRA). The S was then calculated independently with the sectional measurement method from the successive diameters measured at equal intervals of one-twentieth (S 20), one-tenth (S 10), one-fifth (S 5) and one-half (S 2) of the total culm length. The S was also quantified geometrically from the total culm length and the diameter at breast height or at base by assuming that the culm form could be approximated by a cone (S DBH and S DAB). The S FRA was compared with each of the computed S values. For S 10, S 5 and S 2, both the mean relative bias (%BIAS) and relative root mean square error (%RMSE) decreased with an increase in the number of measured diameters. The %BIAS and %RMSE of the S 10 were, respectively, comparable and smaller compared to those of S 20. The bias of the S DBH and S DAB suggested that the cone assumption of the culm form was violated. In conclusion, we recommend that the S should be quantified from the successive diameters measured at equal intervals of one-tenth of the total culm length using the sectional measurement method.