The odd fraction [c] in variation partitioning: a case study on fine scale variation in oak forest herb layer
| Authors | |
|---|---|
| Year of publication | 2013 |
| Type | Conference abstract |
| MU Faculty or unit | |
| Citation | |
| Description | After the introduction of Hubbell’s neutral theory of biodiversity, partitioning of ecological variation into spatial and environmental components has become a common practice in many ecological studies. Along with the rediscovered dispersal issues, popularization of the PCNM method as a spatial modelling tool has undoubtedly contributed a great deal to this trend. The variation partitioning is believed to give us an insight into mechanisms structuring species distributions, with environmental control (i.e. niche processes) one the one side and dispersal limitation (i.e. neutral processes) on the other. In this line of reasoning, the proportion of so called pure spatial variation (fraction [c]) is usually related to the importance of dispersal processes whereas the environmental fraction ([a + b]) is ascribed to niche filtering. It is then perfectly justified to only collect and compare percentages across communities. Nevertheless, what we regard as a more challenging approach is to subject the popular assumptions to closer scrutiny. Here, we focus on the assumption that the pure spatial variation represents the strength of dispersal processes. If it was correct, there should be a link between individual species’ dispersal traits and pure spatial patterns. In addition to it, we examine the role of neglected environmental variables that can, when incorporated into analysis, considerably reduce the proportion of the fraction [c]. To accomplish these goals, we conducted a fine scale study of understory forest vegetation. We collected data from three plots placed in oak and oak-hornbeam forests located in the south-eastern part of the Czech Republic, each plot represented by a set of 100 4m2 subplots organized in a square grid of 1 ha size. The results indicate that on a fine scale, the link between the fraction [c] and dispersal traits is weak and what seems to prevail is the role of environmental variables. |
| Related projects: |