云南红豆杉单木生物量模型选择与树冠调控

拟合并选择云南红豆杉单木生物量模型，明确树冠结构调控技术。以3年生福建省明溪县云南红豆杉原料林为对象，采用4种自变量类型模型，并导入树冠结构特征变量拟合生物量模型；经配对t检验、偏差检验与估算预测精度，优选系列单木生物量模型。比较导入树冠结构特征前后模型的拟合与预测效果，明确并验证树冠结构调控技术。结果表明，采用D²H变量拟合的模型效果最优，其次为DH变量，然后是D²变量，最后是D、H变量。采用D²H变量较D、H变量明显提高模型的拟合与预测效果。导入树冠结构特征后明显提高模型拟合与预测效果。冠幅、冠形率显著影响红豆杉系列单株生物量，促进系列生物量的树冠结构调控重点与方向是促进冠幅宽大；经验证，树冠结构调控技术显著提高系列单木生物量是可行的。自变量类型显著影响模型的拟合与预测效果，选择自变量类型以优化红豆杉单木生物量模型是必要和可行的。导入树冠结构特征因子明显提高单木生物量模型的拟合效果与预测精度。选出4个最优红豆杉单木生物量模型，其决定系数不小于0.862，预测精度均不小于95.18%，M_APE均不大于16.95%，可以用于生产中的生物量预测。

Selection of Single Tree Biomass Model and Canopy Structure Regulation of Taxus yunnanensis

［Objectives］ To fit and select the single tree biomass model of Taxus yunnanensis,and to elucidate the regulation technology of the tree canopy structure.［Methods］ Taking the 3-year-old T.yunnanensis forest in Mingxi County,Fujian Province as the research object,four independent variable type models were adopted,and the canopy structure characteristic variables were introduced to fit the biomass model.A series of single tree biomass models were optimized by paired t-test,deviation test and estimation of prediction accuracy.The fittings and prediction effects of the models before and after the introduction of the characteristics of canopy structure were compared,and the regulation mechanism of canopy structure was revealed and verified.［Results］ The model with D²H variable fitting had the best effect,followed by the one with DH variable fitting,then D² variable fitting,and finally D and H variables fitting.Using D²H variable significantly improved the fitting and prediction effect of the model compared to those with D and H variables.After incorporating the characteristics of canopy structure,the fitting and prediction effect of the model were substantially improved.The canopy breadth and canopy shape rate had significant impacts on the single tree biomass of T.yunnanensis series.The emphasis of canopy structure regulation to improve the series biomass was to increase the canopy breadth.It was proved that a notable increase of single tree biomass of T.yunnanensis series could be achieved by applying canopy structure regulation technology.［Conclusion］ The type of independent variables significantly affects the fitting and prediction effect of the model.It is necessary and feasible to select the type of independent variables to optimize the single tree biomass model of T.yunnanensis.It is of great importance to introduce the characteristics of canopy structure to substantially improve the fitting effect and prediction accuracy of single tree biomass model.Four optimal single tree biomass models of T.yunnanensis were selected.The determination coefficients were not less than 0.862,the prediction accuracies were not less than 95.18%,and the M_APE were not more than 16.95%,which could be used for biomass prediction in production.In order to enhance the single tree biomass of T.yunnanensis,the critical measure of canopy structure regulation is to increase the canopy breadth.