受季风影响的河流底栖硅藻集合群落时间动态

Temporal Dynamics of Benthic Diatom Metacommunities in Monsoon-Influenced Streams

集合群落(Metacommunity)由局地过程和区域过程共同构建和维持。季风区河流水文因子的时间动态对河流集合群落结构和构建方式具有重要影响。自2004年9月至2006年9月,在香溪河调查12个代表性样点,基于集合群落的理论框架,以河流底栖硅藻为研究对象,采用时空交互模型(STI)、集合群落结构组分分析(EMS)、方差分解(Variation Partitioning)等方法,分析河流集合群落结构及构建过程的时间动态,并通过回归分析,探究集合群落时间动态与水文因子的关系。结果表明:河流集合群落结构和构建方式具有明显的时间动态,枯水期和平水期,集合群落分别受扩散限制和环境筛选过程的影响呈现出嵌套结构;丰水期,局地群落间物种的过高扩散产生群块效应,使集合群落呈现出类似Clementsian或Gleasonian的结构。洪水对于河流集合群落构建具有重要作用,7日内的最大流量、群落重置时间,对集合群落构建中空间因子的相对重要性有较为显著的负影响;群落重置时间与集合群落结构呈较为显著的二项式关系,洪水爆发75 d后集合群落开始恢复。

Metacommunities are structured and maintained by a combination of local and regional processes. In monsoon region, the temporal dynamics of hydrological factors have important influences on stream metacommunity structures and structuring. Based on the theoretical framework of metacommunity, taking stream benthic diatoms as research objects, 12 sites in Xiangxi River were investigated from September 2004 to September 2006, the space-time interaction model, elements of metacommunity structure, and variation partitioning were used to analyze the temporal dynamics of the metacommunity structure and structuring processes, and through regression analyses, the relationship between the temporal dynamics of metacommunities and hydrological factors were explored. The results showed that the structures and structuring of stream metacommunity have obvious temporal dynamics. During the dry season and the normal season, the metacommunities presented nested structure, under the influences of dispersal limitation and environmental sorting processes, respectively. The high rate dispersal produced mass effect, which made metacommunity structures similar to Clementsian or Gleasonian. Floods played important roles in structuring stream metacommunity. The maximum flow within 7 days(Q7-max), and the time since three times the median discharge(Reset time) had significant negative impacts on the relative importance of spatial factors in metacommunity structuring. The reset time and the metacommunity structures showed a slightly significant binomial relationship, the metacommunity began to recover from 75 days after the flood.