基于相邻木排列关系的混交度研究

混交度是反映森林群落中树种相互隔离状况的一个重要指标,目前已提出多种混交度指数.Mg考虑了对象木与最近相邻木的树种异同,Mv和Ms进一步考虑了最近相邻木(空间结构单元)树种数,但仍不能准确描述树种隔离程度.基于相邻木排列关系的混交度Mp,利用“1+4”结构的17个不同空间结构单元和天目山常绿阔叶林数据对Mg、Mv、Ms和Mp4个混交度进行了比较分析.结果表明:Mp的区分能力最强,Mv和Ms次之,Mg最差.Mp最适用于分析混交结构复杂的林分.在天目山常绿阔叶林分析中,混交度水平为中度,4个最近邻体中有2个相同种顺序排列的空间结构单元个数明显多于交错排列的个数.基于相邻木排列关系的混交度Mp能准确区分树种隔离程度,提高了混交度的区分度,可反映森林实际混交状况.

Study of mingling based on neighborhood spatial permutation

Mingling, its concept was put forward by Gadow in 1992, is an important indicator to reflect species mutual segregation in a forest community, and is a significant forest spatial structure index. Currently, many mingling indices have been proposed. A simple mingling index Mg proposed by Gadow only involves the difference between objective trees and their nearest neighbor trees, and cannot accurately reflect the response of tree species to spatial segregation, which was noticed by Tang Mengping(2004) and Hui Gangying(2008). They put forward tree species diversity mingling Mv (further considering the number of nearest tree species) and tree species spatial status Ms (further considering the number of tree species in spatial structure unit), respectively. However, Mv and Ms cannot distinguish the difference between two of the same species on four of the nearest tree species and three of the same species on four of the nearest tree species. Actually, two of the same species on four of the nearest tree species are different in sequence and interleaving arrangement, but they are the same in the analysis of Mv and Ms. Virtually, neighborhood spatial permutation is a significant factor in describing tree species spatial segregation, which was ignored by Mv and Ms. Therefore, a new mingling index, Mp based on neighborhood spatial permutation has been put forward in this paper."1+4" structure, which was the abbreviation of optimum spatial structure unit proposed by Hui Gangying(2003), composed by one objective tree and four of its nearest trees, and has 17 spatial structure units that are different in tree species spatial permutation. Mp was compared with Mg, Mv and Ms through 17 spatial structure units and evergreen broad-leaved data from Tianmu Mountain, a national nature reserve in Zhejiang province. Data were collected in a 1hm²(100m×100m) permanent plot that had been established in 2005. The results show that Mp has the strongest distinction among them, Mv and Ms are the second, Mg is the worst. The ranking of the different mingling values was: Mg>Mp>Ms, Mg>Mv. Ms and Mp are both able to adequately analyze the simple mixed stand. Mp is the best mingling index among them to analyze the complex mixed stand. Here, there is a high value of mingling, if Mg is used. Sometimes, Mv is unable to analyze the tree species spatial segregation. Spatial structure units which have two of the same species on four of the nearest tree species are different in sequence and interleaving arrangement, and the number of sequence arrangement is significantly more than that of interleaving, and thus there is a moderate mingling level on evergreen broad-leaved in Tianmu Mountain. Mp is based on neighborhood spatial permutation and can distinguish tree species spatial segregation accurately, and has higher distinction degree accuracy than that of traditional mingling. This better reflects the actual mixed situation of forest, and is an effective indicator to reflect species mutual segregation in a forest community. "1+4" structure has 17 different spatial structure units, however, what needs further study is the number of spatial structure units in the "1+n" structure, and how to determine the deduction relation. Further, the number of units may obey certain mathematic distributions in natural mixed forests, and the mathematic distribution function may be a significant reference to measure the near-natural degree of cultivation in mixed forests. These issues need further study.