云南松单木材积生长率模型研究

Study on the Volume Growth Rate Model of Pinus yunnanensis of Individual Tree

基于云南省第六次(2002年)和第七次(2007年)森林资源连续清查数据,以云南省滇中地区为研究区,采用非线性回归模型,建立不同海拔、起源、密度和龄组的单木材积生长率模型。以决定系数(R2)和均方根误差(RMSE)作为最优模型选取标准,以总体相对误差(RS)、平均相对误差(EE)、绝对平均相对误差(RMA)和预估精度(P)作为最优模型精度检验指标。结果表明:滇中地区云南松单木材积生长率随胸径的增加呈反“J”型曲线分布,多属幼龄林、小径木;海拔、起源及密度等指标下云南松单木材积生长率模型拟合精度较高,模型参数稳定,最优模型决定系数R2均在0.8以上,模型预估精度除龄组中过熟林之外,其余均在80%以上;过熟林木因仅有15株数据样本,模型预估精度只达到70%;整体上拟合效果良好。模型拟合效果好、适用性强,可用于云南松材积生长量的估算、编制云南松材积生长率表及对未来云南松林分资源的动态变化进行预测。

Based on the data of the sixth and seventh Continuous Forest Inventories(CFI),this paper used the nonlinear regression model to establish the volume growth rate model of Pinus yunnanensis of individual tree with different altitudes,origin,age classes and stand density.Compared to the optimal model in the case of each index by determination coefficient and root mean square error,the model precision was inspected by sum relative error,mean relative error,absolute mean relative error and predict precision.The result showed that the volume growth rate decreased with the increase of diameter.The fitting precision of the volume growth rate model of Pinus yunnanensis of individual tree was high with different altitudes,origin,age classes and stand density.The model parameters were stable and the determination coefficient of optimal models was more than 0.8.The overall model prediction precision was more than 80 percent except for the over-mature forest with seventy percent due to the less data.The fitting precision of optimal model was high and the applicability was strong.The optimal volume growth rate models can be used to estimate the volume growth,compile the volume growth rate table and forecast the dynamic changes of Pinus yunnanensis forest resources,and provide reference value for the forest resources inventory.