西双版纳地区六种林型地表蜘蛛多样性比较研究

为探索地表蜘蛛多样性及其变化与森林类型和管理方式的关系, 在西双版纳勐仑自然保护区选择热带季节雨林、石灰山季节雨林和山地常绿阔叶林, 在自然保护区附近选择人工纯林、胶茶群落和橡胶林, 共6种林型, 每种林型选择3块样带, 共设置研究样地18块, 分别于2006年12月上旬（雾凉季）、2007年3月下旬（干热季）和2007年7月上旬（雨季）, 以单位地表面积法收集地表蜘蛛的物种组成和数量数据, 并以蜘蛛种类和数量分布为属性进行典范对应分析（CCA）, 探讨不同类型植被与地表蜘蛛多样性的关系。共采集蜘蛛标本9 849头, 用于统计分析的成熟蜘蛛3 119头, 归属于30科, 其中幽灵蛛科、皿蛛科、球蛛科和小密蛛科是地表蜘蛛的优势类群。各林型科的数量为: 热带季节雨林24科, 石灰山季节雨林22科, 山地常绿阔叶林22科, 人工纯林20科, 胶茶群落21科, 橡胶林19科; 各林型特有科数量: 热带季节雨林2科, 山地常绿阔叶林2科, 橡胶林1科; 而仅在雨林中分布的科4个（占全部30个科的13.3%）, 仅在自然林中分布的科6个（20.0%）, 仅在人工林中分布的科1个（3.3%）。从蜘蛛的数量分布看, 个体密度在热带季节雨林显著高于其他5种林型; 橡胶林多样性指数和丰富度指数显著低于3种自然林, 而均匀度指数的最低值也同样在橡胶林出现。CCA分析和聚类分析的结果表明, 6种林型趋于分成2组, 即: 自然林和人工林; 在自然林中两种次生林的相似程度更高; 人工林中人工纯林（非橡胶林）与胶茶群落的相似程度更高。以上结果表明: （1）森林砍伐后种植人工林措施改变了该地区地表蜘蛛群落的物种分布格局; （2）蜘蛛多样性随着人为干扰程度增加有减少的趋势; （3）减少人为干扰和增加植被群落多样性（橡胶林中种植茶树）对保护和恢复物种多样性有重要意义。

Biodiversity of ground-dwelling spider in six forest types in Xishuangbanna, S.W. China

The composition and distribution of ground-dwelling spider were investigated using small plot direct searching method to assess the impact of forest type and habitat management on spider diversity. Six forest types in Xishuangbanna, Yunnan Province, southwestern China were surveyed, i.e., the tropical seasonal rain forest, limestone tropical seasonal rain forest, tropical montane evergreen broad-leaved forest, artificial forest, rubber-tea community and rubber plantation. A total of 18 plots (three repetitions for each forest type) were surveyed during three field trips in December 2006, March and July 2007. The relationships between forest type and spider diversity were explored using canonical correspondence analysis (CCA). A total of 9 849 individual spiders were collected from various habitats, of which 3 119 are identifiable adults representing 30 families. The most abundant families of ground-dwelling spider are Pholcidae, Linyphiidae, Theridiidae and Mysmenidae (62.2% of total species counts). Of the six forest types, 24 families were collected from the tropical seasonal rain forest including two endemic families; 22 families were collected from the limestone tropical seasonal rain forest; 22 families were collected from the tropical montane evergreen broad-leaved forest including two endemic families; 20 families were collected from the artificial forest; 21 families were collected from the rubber-tea community; and 19 families were collected from the rubber plantation including one endemic family. There are four families (13.3% of the total 30 families recorded) endemic to two types of rain forest, six families (20.0%) endemic to three types of natural forest, and one family (3.3%) endemic to plantations. Abundance of ground-dwelling spider in the tropical seasonal rain forest is significant higher than the remaining five types of forest. Shannon-Wiener (H′) index and Margalef index (Dmg) in the rubber plantation are significant lower than the three types of natural forest, and Pielou index (J) is also the lowest in the rubber plantation. The results of CCA and cluster analysis indicated that natural forests and plantations are well differentiated. In the group of natural forests, the tropical seasonal rain forest plots are well separated from two types of secondary forest plots. In the other group, the rubber plantation plots are well separated from the artificial forest and the rubber-tea community plots. The results suggested that: (1) human-management has an strong impact on ground-dwelling spider composition; (2) spider diversity decreases along with the increase of anthropogenic disturbance; and (3) reducing anthropogenic disturbance and increasing plant community diversity (e.g., plant tea tree in rubber plantation) is a meaningful way to conserve and restore biodiversity.