Negative relationship between interspecific spatial association and trait dissimilarity

25 October 2018

He, Dong; Biswas, Shekhar

Species’ response to environmental site conditions and neighborhood interactions are among the important drivers of species’ spatial distributions and the resultant interspecific spatial association. The importance of competition to interspecies spatial association can be inferred from a high degree of trait dissimilarity of the associated species, and vice versa for environmental filtering. However, because the importance of environmental filtering and competition in structuring plant communities often vary with spatial scale and with plant life stage, the species’ spatial association – trait dissimilarity relationship should vary accordingly. We tested these assumptions in a fully mapped 50-ha subtropical evergreen forest of China, where we assessed the degrees of interspecies spatial associations between adult trees and between saplings at two different spatial scales (10 m versus 40 m) and measured the degrees of trait dissimilarity of the associated species using six traits (leaf area, specific leaf area, leaf dry-matter content, wood density, wood dry-matter content and maximum height). Consistent across spatial scales and plant life stages, the degree of interspecific spatial association and the degree of overall trait dissimilarity (i.e., all six traits together) was negatively correlated, suggesting that environmental filtering might help assemble functionally similar species in our studied communities. However, when we looked into the spatial association – trait dissimilarity relationship for individual traits, we found that the relationships between interspecific spatial associations and the dissimilarity of wood density and dry-matter content were significant for adults but not for saplings, suggesting the importance of wood traits in species’ survival during ontogeny. We conclude that processes shaping interspecific spatial association are spatial scale and plant life stage dependent, and that the distributions of functional traits offer useful insights into the processes underlying community spatial structure.