全尺度人工湿地中植物多样性对生产力与多样性效应的影响

在中国东南部的全尺度复合垂直流人工湿地中开展2年的植物多样性实验，以研究植物多样性（包括植物物种丰富度和植物组成）对群落生产力与多样性效应（即互补效应、选择效应和净多样性效应）的影响及其产生机制。结果表明，2007年物种丰富度与群落生产力呈线形正相关，而2008年显著的单峰格局，其关系式为：y=-0.213x2＋3.455x＋15.192（R=0.215）。2008年物种丰富度与互补效应呈显著地线形负相关，而2007年呈单峰格局，其关系式为：y=-0.389x2＋6.974x-10.707（R=0.247），而且2007年与2008年的互补效应与生产力都呈显著的正相关，表明互补效应对生产力的提高有重要作用。然而，2007年与2008年物种丰富度与选择效应之间均没有显著相关性，且选择效应与群落生产力之间也没有显著相关性，表明选择效应对生产力的提高作用不显著。2007年与2008年中物种组成对生产力、互补效应、选择效应与净多样性效应均有显著影响，说明人工湿地的植物配置对其生态系统功能的维持尤为重要。2008年物种丰富度与净多样性效应呈极显著地线形负相关，而2007年呈显著单峰格局，其关系式为：y=-0.329 x2＋5.968 x-12.659（R=0.234），这种趋势主要是由于植物多样性-生态系统功能关系的影响因素（如物种的竞争力和生态位）在2年中有所变化。同时，2007年与2008年的多样性净效应与生产力都呈显著正相关关系，表明生产力与多样性净效应的变化趋势是同步的。与抽样效应假说不同的是，本实验中单种最高产物种（芦竹）在混种时没有表现出高产，主要是由于生长的分配、资源的竞争力与环境的变化等。

Effects of plant diversity on productivity and diversity effect in a full-scale constructed wetland

An experiment was conducted in a full-scale constructed wetland in Southeastern China between 2007 and 2008 to study the effects of plant diversity （including plant species richness and plant compositions） on the community productivity （such as complementary effects, selection effects and net diversity effects）. The results show a strong, positive correlation between the community productivity and plant species richness in 2007. With a unimodal pattern, the community productivity correlated poorly with the species richness in 2008 （R=0.215 for the fitted relationship of y=-0.213x2＋3.455x＋15.192）. This indicates that the interaction of environmental conditions and competition played an important role in the formation of the above relationships. The plant species richness was found to correlate negatively with complementary effects in 2008, but a unimodal pattern was evident in 2007 （R=0.247 for the fitted relationship of y=-0.389 x2＋6.974 x-10.707）. Meanwhile, a significant, positive relationship between complementary effects and productivity show that complementary effects played an important role in the improvement of community productivity in both 2007 and 2008. In contrast, no significant relationship found among selection effects and plant species richness and productivity, suggesting that selection effects had no important role in the improvement of community productivity in 2007 and 2008. Plant composition had significant effects on the productivity as well as complementary effects, selection effects and net diversity effects in 2007 and 2008. This indicates that the plant configuration in the full-scale constructed wetland was particularly important for maintaining the functionality of the ecosystem. In addition, the plant species richness was found to correlate negatively with net diversity effects in 2008, but such a correlation was absent in 2007 and instead a unimodal pattern was evident （R=0.234 for the fitted relationship y=-0.329 x2＋5.968 x-12.659）. The different correlation patterns were due to variations of various factors that influenced the biodiversity-system functioning in the two years, such as competitiveness of different species and ecological niche. No consistent relationship between selection effects and productivity was found, suggesting that the variation trends of net diversity effects and productivity were synchronous. Inconsistent with the sampling effect hypothesis, the species （Arundo donax） with the highest monoculture yield did not produce higher yield in the mixtures, probably due to growing allocation, resources competition and environmental changes. The results from this study provide data support to the research on and application of the relationship between plant diversity and ecosystem functioning in constructed wetland experiments, and also offer a theoretical base for scientific management of constructed wetland systems.