基于信息熵的森林立地离散空间场相关性研究

为了获取立地空间信息间隐含的相关性，对其进行定量计算和表达，研究选取森林资源小班调查数据中7项典型的离散数据，结合小班空间位置属性构成森林立地离散空间场，采用信息熵的理论方法，通过计算局部空间内离散场的信息量以及局部空间与整体的协调性，定量分析并提取多项离散型因子与立地森林健康等级间的相关指数。结果表明，在7项因子中不同的立地类型和小班内的优势树种与森林健康相关程度最高，森林起源则与森林健康等级表现出相互独立的关系。研究克服了以往使用统计学原理以及灰色系统理论均无法计算立地离散空间场相关性的缺陷，实现了对立地因子中的离散型属性间关系的定量计算和表达。

Information Entropy Based Study on the Correlations of Discrete Spatial Fields of Forest Sites

The attributes of forest site space information are mostly identified by classification codes, therefore, it is unable to calculate their correlations by mean value and variance. In order to obtain the implied correlations among these attributes, they were calculated and expressed quantitatively. From forest sub-lot data, combining with sublot space location to form forest discrete spatial fields, 7 typical discrete attributes were chosen. The research used theoretical method of information entropy, via calculating the information quantity inside partial space, to quantitatively analyze and extract the correlation indices between discrete factors and health levels of the site forests. The results suggested that different site types and dominant tree species revealed the highest correlation with forest health, while the origin of forest and health level of forest showed an independent relationship. The study overcomes the defects that site discrete spatial correlation cannot be calculated by statistics theory and grey system theory, eventually achieved quantitative calculation and express of the relationship between site discrete attributes.