Appendix D. Comparison of the ability of three different metrics of even trait spacing to distinguish between alternative models of community assembly.
|
|
| FIG. D1. A representative example of the ability of three even spacing metrics to distinguish between the patterns produced by different models of community assembly. In each panel, the median and the 5 and 95% quantiles of distribution of trait metric values produced by the random assembly model is shown in gray, the habitat filtering model is shown in blue, and the competition model is shown in red. The example shown is for D950.1 (mm, log10 transformed), a proxy for maximum height, at the 20 m scale. In panel A, the standard deviation of nearest neighbor distances (SDNN) is more sensitive to the reduced range of traits produced by the habitat filtering model than to the even spacing of species produced by the competition model, as others have cautioned (Stubbs and Wilson 2004, Cornwell and Ackerly 2009). Dividing this metric by the observed range of values within a community (SDNNr, panel B)(Stubbs and Wilson 2004, Kraft and Ackerly 2009) offers some improvement in that the habitat filtering and competition models overlap more, and, at lower species richness levels, filtering is more similar to random than the competition model. The standard deviation of all neighbor distances divided by range (panel C, SDNDr, see methods) (Ingram and Shurin 2009) appears to offer more improvement still, in that the random and habitat filtering models overlap considerably, while the competition model produces smaller SDNDr values than the other two models for about half of the range of richness levels in the forest. There is, however, still room for improvement, as even with SDNDr the median of the habitat filtering model is slightly reduced relative to the median of the random model, and at higher richness levels all three assembly models converge on similar values. |
LITERATURE CITED
Cornwell, W. K., and D. Ackerly. 2009. Community assembly and shifts in the distribution of functional trait values across an environmental gradient in coastal California. Ecological Monographs 79:109–126.
Ingram, T., and J. B. Shurin. 2009. Trait-based assembly and phylogenetic structure in northeast Pacific rockfish assemblages. Ecology 90:2444–2453.
Kraft, N. J. B., and D. D. Ackerly. 2009. Response to comment on "Functional traits and niche-based tree community assembly in an Amazonian forest". Science 324:1015d.
Stubbs, W. J., and J. B. Wilson. 2004. Evidence for limiting similarity in a sand dune community. Journal of Ecology 92:557–567.