基于不同零模型的三江平原沼泽湿地主要物种小尺度点格局分析

在三江平原沼泽湿地,基于不同零模型的点格局方法,研究了常年积水环境条件下沼泽湿地主要优势种毛苔草、漂筏苔草、狭叶甜茅以及小叶章种群的空间分布格局特征。结果表明:在0—200cm尺度范围内4种物种基本都偏离完全随机模型,表现为聚集分布,但偏离的程度不同;除了数量最少的小叶章种群外,其他3种物种都在一定尺度上偏离了泊松聚块模型;毛苔草和狭叶甜茅种群在极小尺度上偏离嵌套双聚块模型,但不显著,而漂筏苔草种群在所有尺度上符合嵌套双聚块模型。种群的实测值偏离完全随机模型的程度越大,越有可能符合符合泊松聚块模型,偏离泊松聚块模型的程度越大,越有可能符合嵌套双聚块模型。4种物种在0—200cm尺度范围内形成的聚块是由营养繁殖引起的,多个分株系统组成大聚块,而每个分株系统形成一个小聚块,聚块的形成加剧了种内竞争,使得种群的聚集强度降低。

Small-scale point pattern analysis based on different null models for detecting spatial patterns of dominant species in Sanjiang Plain,China

Marsh was the major wetland type. The soil of marsh is oxygen-poor due to permanently inundate which has formed a special environmental conditions and vegetation types. The Sanjiang Plain is one of the largest freshwater marshes with area of approximately approximately 10400km². The dominant plants are grass species which both have cloned habits such as Carex lasiocarpa, Carex pseudocuraica, Deyeuxia angustifolia and Glyceria spiculosa. To study the spatial distribution pattern of plant populations in marsh is supplementary for spatial pattern theory because the distinctive environmental conditions. Meanwhile, it is very significance for revealing plant populations characteristics in marsh environment. Thus, we analysis the spatial distribution pattern of main dominant species (C. lasiocarpa, C. pseudocuraica, D. angustifolia and G. spiculosa) by point pattern analysis based on different null models. The study site is located at the Ecological Experiment Station of Mire Wetland in Sanjiang Plain, Chinese Academy of Sciences (47°35''N,133°31''E, 56m above sea level), Northeast of China. In September 2012, we mapped a 4×4 m representative area in the permanently inundated marsh community which the surface are smooth and have uniform physiognomy. The plot was divided into 256 contiguous 25×25 cm quadrats, as the basic unit of vegetation survey, using the bamboo pole and the white rope. The relative position of each plant individual or plexus projection of grass was expressed in the coordinates which represent the distance. All analyses were conducted using the grid-based estimators in the Programita software package. Nineteen Monte-Carlo simulations were used to generate 95% confidence envelopes. Our results showed that: The four main dominant species mainly depart from complete spatial randomness (CSR) and shows aggregation in 0-200cm scales, but have varying degrees of deviation; The C. lasiocarpa, C. pseudocuraica and G. spiculosa depart from poisson cluster process (NS) in small-scale, except D. angustifolia; The C. lasiocarpa and G. spiculosa depart from nested double-cluster process (DC) in a very small-scale, but not significant. The C. pseudocuraica accordance with nested double-cluster process in 0-200cm scales, completely; The measured value of population farther departed from complete spatial randomness (CSR) are more likely accordance with poisson cluster process (NS), and farther depart from poisson cluster process (NS) are more likely accordance with nested double-cluster process (DC); The aggregation of 4 species in 0-200cm scales is caused by vegetative propagation which could help to adapt the perennial water environment. Much ramet system can form a cluster in large scale, while the one ramet system forms a cluster in small scale. On the other side, the aggregation intensity of population would reduce due to strong intraspecific competition in has formed cluster.