Department of Biology, Syracuse University, 114 Life Sciences Complex, Syracuse, NY 13244, USA.
Oecologia. 2012 Nov;170(3):767-75. doi: 10.1007/s00442-012-2351-7. Epub 2012 May 13.
Environmental filtering and niche differentiation are processes proposed to drive community assembly, generating nonrandom patterns in community trait distributions. Despite the substantial intraspecific trait variation present in plant communities, most previous studies of trait-based community assembly have used species mean trait values and therefore not accounted for intraspecific variation. Using a null model approach, I tested for environmental filtering and niche differentiation acting on three key functional traits--vegetative height, specific leaf area (SLA), and leaf dry matter content (LDMC)-in old-field plant communities. I also examined how accounting for intraspecific variation at the among-plot and individual levels affected the detection of nonrandom assembly patterns. Tests using fixed species mean trait values provided evidence of environmental filtering acting on height and SLA and niche differentiation acting on SLA. Including plot-level intraspecific variation increased the strength of these patterns, indicating an important role of intraspecific variation in community assembly. Tests using individual trait data indicated strong environmental filtering acting on all traits, but provided no evidence of niche differentiation, although these signals may have been obscured by the effects of dispersal limitation and spatial aggregation of conspecific individuals. There was also strong evidence of nonrandom assembly of individuals within single species, with the strength of environmental filtering varying among species. This study demonstrates that, while analyses using fixed species mean trait values can provide insights into community assembly processes, accounting for intraspecific variation provides a more complete view of communities and the processes driving their assembly.
环境过滤和生态位分化是驱动群落组装的过程，它们会在群落特征分布中产生非随机模式。尽管植物群落中存在大量的种内特征变异，但基于特征的群落组装的大多数先前研究都使用了物种平均特征值，因此没有考虑到种内变异。我使用零模型方法，测试了环境过滤和生态位分化对三种关键功能特征——营养体高度、比叶面积（SLA）和叶干物质含量（LDMC）——在旧农田植物群落中的作用。我还研究了在个体和个体间水平上考虑种内变异如何影响非随机组装模式的检测。使用固定物种平均特征值的测试提供了证据表明，环境过滤作用于高度和 SLA，生态位分化作用于 SLA。包括样方水平的种内变异增加了这些模式的强度，表明种内变异在群落组装中起着重要作用。使用个体特征数据的测试表明，所有特征都存在强烈的环境过滤，但没有证据表明存在生态位分化，尽管这些信号可能被扩散限制和同种个体的空间聚集的影响所掩盖。在单个物种内个体的非随机组装也有很强的证据，环境过滤的强度在物种间存在差异。这项研究表明，虽然使用固定的物种平均特征值进行分析可以提供对群落组装过程的深入了解，但考虑种内变异可以提供更全面的群落视图以及驱动其组装的过程。
