福建杉木树冠外轮廓和树冠体积相容性模型

Crown profile and volume compatibility model of Cunninghamia lanceolata in Fujian Province

[目的] 树冠外轮廓模型不仅能够描述任意位置处树冠半径,而且能够推导预测树冠体积与树木地上部分生物量。根据福建地区98块杉木Cunninghamia lanceolata人工林样地的413株杉木调查数据,构建了具有同一套模型参数的树冠外轮廓模型和树冠体积预测联立方程组系统。 [方法] 选取4种常用的可积分树冠外轮廓备选模型,利用积分法对备选模型进行推导,得到树冠体积预测模型;将模型方程分别两两联立建立树冠外轮廓与体积相容性联立方程组,并利用SAS软件模块中的似乎不相关回归过程估计联立方程组模型系统的参数。为了消除模型异方差,采用加权回归方法拟合模型,并对不同模型系统的拟合精度、预测精度进行对比分析。 [结果] 基于模型4的联立方程组拟合精度高、预测性能好,其树冠外轮廓和体积拟合精度分别达到0.829 5和0.861 0,预测结果精度分别为0.803 9和0.856 0;通过似乎不相关回归法解决了联立方程组共线性问题,加权回归方法一定程度上消除了模型中存在的异方差性。 [结论] 所构建的树冠外轮廓-体积一致性模型方程实现了树冠外轮廓与体积模型之间互相推导,为进一步估测树木地上部分生物量提供了理论依据。

Objective Crown profile models can not only describe the radius at any position of the crown, but also predict the crown volume and the aboveground biomass. This paper developed a simultaneous equation system of crown profile model and crown volume prediction with the same set of model parameters using the data of 413 trees in 98 sample plots of Cunninghamia lanceolata plantation in Fujian Province. Method Four commonly used and integral crown profile models were selected, and the crown volume prediction models were derived by using integral method. The model equations were set up in pairs to establish the simultaneous equations of tree crown profile and volume compatibility, and the seemingly unrelated regression (SUR) process in SAS software module was used to estimate the parameters of the simultaneous equations model system. In order to eliminate the heteroscedasticity of the model, the weighted regression method was used to fit the model. The fitting accuracy and prediction accuracy of different model systems were compared and analyzed. Result The simultaneous equations based on model 4 had high fitting accuracy and good prediction performance. The fitting accuracy of the crown profile model and crown volume model reached 0.829 5 and 0.861 0 respectively, and the accuracy of prediction was 0.803 9 and 0.856 0 respectively. The collinearity problem of simultaneous equations was solved by the SUR method, and the heteroscedasticity in the model was eliminated to some extent by the weighted regression method. Conclusion The consistency model equation of crown profile-volume constructed in this paper can be used to derive the crown profile and volume model, which provides a theoretical basis for estimating the biomass of the aboveground part of trees.