论文部分内容阅读
对用于评价印度拉贾斯坦邦热带干旱区沙漠柚木极限林分密度的公式进行了研究。对5个不同密度的林分基面积预测模型进行测试和比较,这些模型属非线性生长函数的轨迹不变量代数差形式。这些模型可根据林分变量,如优势高/公顷、立木数/公顷,预测林分基面积,还可用于评估不同造林方案。利用所收集的22块样地的数据建立模型。使用各种可能的树龄数据结构,用定性和定量标准比较了这些替代模型。使用Akaike信息准则差异统计来分析各种模型的预测能力。结果表明:Hui和Gadow提出的模型预测效果最好,推荐在本研究区域内使用该模型来预测沙漠柚木林分基面积。但是,此模型并未使用间伐林分数据,因此,不能准确地预测林木有明显自然死亡的林分基面积。图3表4参28。
A study was conducted on the formula used to evaluate the density of the limiting tree stand of desert teak in the tropical arid region of Rajasthan, India. Five different densities of forest base area prediction models were tested and compared. These models belong to the algebraic form of trajectory invariance of nonlinear growth functions. These models can be used to assess different afforestation scenarios based on stand variables such as dominance / ha, stand / ha, projected stand area. The collected data from 22 plots were used to establish the model. Using a variety of possible tree-age data structures, these alternative models were compared using qualitative and quantitative criteria. Akaike Information Criterion difference statistics were used to analyze the predictive power of various models. The results show that the prediction model proposed by Hui and Gadow is the best, and it is recommended to use this model to predict the area of desert teak in this study area. However, this model does not use the thinning stand data, therefore, it is not possible to accurately predict the stand area of the tree with obvious natural death. Figure 3 Table 4 参 28.