Using allometric equations to predict biomass is a reliable approach for estimating biomass in bamboo forests. This approach has been widely applied in various bamboo forests worldwide. However, the independent variables used in allometric models are inconsistent across bamboo species. Therefore, selecting the appropriate independent variables for a specific species is essential for predicting biomass. The present study addressed a Makino bamboo (Phyllostachys makinoi) plantation. A total of 81 samples were employed in this study to develop aboveground biomass (AGB) models. The datasets contained diameter at breast height (DBH), culm height (H), age (A), foliage biomass, branch biomass, and culm biomass for each sample. We used AGB as a dependent variable and DBH, H, and A as independent variables to develop four models. Each model contained one to three independent variables. Four indicators, R2adj, the residual sum of squares, root mean square error, and Akaike information criterion, were employed to examine the models. The findings demonstrated that only using DBH had an excellent performance for AGB prediction, as AGB = 0.435 & times; DBH1.621. It indicated that adding other variables did not significantly promote the predictive effects for the models. This study further adopted DBH to predict the biomass of foliage, branches, and culms. The results showed that the model used in culms had the best performance due to the highest R2adj. Therefore, our study suggested that using only DBH as a predictor for AGB prediction was adequate and recommended it for Makino bamboo.
原文链接
分享
引用
