Intraspecific variation in traits reduces ability of trait‐based models to predict community structure. (4th August 2017)

• Record Type: Journal Article
• Authors: Read, Quentin D.; Henning, Jeremiah A.; Sanders, Nathan J.
• Editors: Mason, Norman
• Is Part Of: Journal of vegetation science. Volume 28:Number 5(2017:Sep.)
• Page Start: 1070
• Page End: 1081
• Publication Date: 2017-08-04
• Subjects: Community model -- Functional trait -- Meadow -- Montane -- Niche -- Plant -- Sub‐alpine -- Traitspace; Plant ecology -- Periodicals; Plant communities -- Periodicals; Plant populations -- Periodicals; 581.7
• DOI: 10.1111/jvs.12555
• Languages: English
• ISSNs: 1100-9233

Abstract

Abstract: Questions: Is it possible to predict the composition of local plant assemblages? Trait‐based approaches have offered some promise, especially in cases where deterministic processes such as environmental filtering and niche differentiation shape communities. In this study, we asked how much intraspecific variation contributes to trait distributions within and among plant communities, and whether trait distributions resulting from environmental filtering and niche differentiation can predict accurately the relative species abundances of montane plant species in local communities. Location: West Elk Mountains, Colorado, USA. Methods: We collected functional trait, species composition and environmental data from 14 sites along a broad climate gradient in Colorado, USA, ranging in elevation from 2480 to 3560 m. We partitioned the variation within and among sites into intraspecific and interspecific components, and compared the results to values from a recent global meta‐analysis, which examined intraspecific trait variability patterns. We also used these data to parameterize statistical models that have been shown to reproduce patterns associated with the processes of environmental filtering and niche differentiation. We fit two models to the data, one assuming that niche differentiation is invariant among sites, and another assuming that niche differentiation varies among sites. Results: We found that the trait‐based models were worse at predicting species relative abundances in local communities than a null model assuming equal abundances of all species. One plausible explanation for the poor performance of the models is that intraspecific variation in functional traits, which in our system was higher than the global averages documented in the meta‐analysis, swamped the effects of interspecific variation in functional traits along the climatic gradient. In particular, almost all variation in root traits was within rather than among species, even among sites. Conclusion: Our results suggest that a greater focus be placed on measuring intraspecific trait variability and determining its consequences for community assembly and ecosystem properties. Abstract : For trait‐based community ecologists, the ability to predict community assembly based on functional traits is the Holy Grail. In this paper, novel hierarchical models parameterized with trait data from plant communities and incorporating individual variation fail to adequately predict community assembly–for these communities, the Holy Grail is still elusive.