树木直径生长的时间序列分析及模型预测  

Rescale Range Analysis and Model Prediction of the Vegetal Process of Tree Diameter

作　　者：罗旭[1] 程承旗[1] 冯仲科[2] 陈晓雪[2] 

机构地区：[1]北京大学地球与空间科学学院 [2]北京林业大学资源与环境学院,北京100083

出　　处：《中南林业科技大学学报》2007年第2期7-12,共6页Journal of Central South University of Forestry & Technology

基　　金：国家"863"计划基金(2006AA706209-2);国家自然科学基金(90302014)

摘　　要：以解析木的数据为基础,对甘肃省小陇山地区的解析木干曲线的分形特征进行了研究,从整体上对树干形状进行描述,采用R/S方法即时间序列分析法对直径的生长过程进行了分析,发现随着树木年龄的增加,其霍斯特指数值也呈现增加的趋势,并以此建立了分形生长的霍斯特(H)指数预测模型,可用于森林生长量的预测,R/S分析结果充分反映了直径动态生长过程随时间尺度变化的分形特征.结合样地及样本数据,以分形分析、线性及非线性回归为主要方法,进而为森林资源的动态监测奠定基础,具有非常重要的生态学意义.Based on the data obtained from parse trees, some research was conducted on the fractal characteristics of the stem curve of parse trees which grow in the district of Xiaolongshan, Gansu Province. The paper describes the figure of the stem as a whole, adopts the R/S method （Rescale Range Analysis） to analyze the vegetal process of the tree diameter, and finds that the Hurst index increases with the age of the trees. Built up on this, a Hurst index forecasting model was set up to be used in the forecast of forest vegetal amount. The result of the R/S analysis fully reflects the fractal characteristics of the dynamic vegetal progress of the diameter along with the change of time scale. Combining sample plots with sample data and taking fractal analysis, linear and nonlinear regression, and error analysis as the main methods, this research establishes a base of dynamic supervision for forest resources, which is of important ecological significance.

关 键 词：森林经理 解析木 R/S分析方法 霍斯特指数 生长预测 

分 类 号：S711[农业科学—林学] S718.45