期刊
JOURNAL OF ECOLOGY
卷 100, 期 3, 页码 577-585出版社
WILEY
DOI: 10.1111/j.1365-2745.2011.01946.x
关键词
competition importance and intensity; indices; plant-plant interactions; productivity gradients; stress; variance decomposition
资金
- NERC [NE/H020802/1, NE/I022027/1] Funding Source: UKRI
- Natural Environment Research Council [NE/H020802/1, NE/I022027/1] Funding Source: researchfish
1. Many studies have explored how plant interactions vary along environmental gradients, and these relationships are fundamental to the many ideas in vegetation dynamics. 2. It has been argued that studies should differentiate between the intensity (absolute effect) and importance (relative effect) of a process, and several important indices have been presented in the literature. 3. We explore the mathematical properties of one of the most popular of these indices (C-imp) and demonstrate that it is (i) insensitive to the underlying ecology, (ii) highly constrained, so making statistical analysis difficult and (iii) can have more complex behaviour than the processes it seeks to summarise. 4. We demonstrate that the positive relationship between C-imp and productivity, which has been used to support the ideas of Grime, is entirely predictable without any information on competition and how it affects plant performance. 5. The properties of three other competition indices (ACI, RCI and lnRR) in relation to habitat productivity are briefly presented. ACI is closely related to C-imp and so shares many of its properties, see Summary point 3 above. 6. An alternative method of analysing variation in plant performance on environmental gradients is then presented. This approach, which is based on a variance decomposition, is easy to implement and has a simple biological interpretation. 7. Synthesis. We strongly discourage the use of heuristic indices in ecological research. The majority of the 50 or so indices that have been used in competition studies have been developed with little or no regard to underlying theory or models. However, it is possible to make predictions about how competition varies along gradients from models that translate into both qualitative and quantitative conditions and metrics, and we encourage authors to do this.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据