COS 15-4 - Do plants display "optimum" traits in optimal habitats?

Tuesday, August 9, 2016: 9:00 AM
315, Ft Lauderdale Convention Center
Rachel M. Mitchell, Biology Department, Duke University, Durham, NC and Justin P. Wright, Biology, Duke University, Durham, NC
Background/Question/Methods

Plant trait expression is strongly shaped by the abiotic environment. Abiotic filters can alter expression of both trait means and variances, potentially leading to strong trait divergence in habitats where species are rare vs. common. We examined whether species demonstrated differing trait expression when growing in habitats that differed in terms of their suitability. We also looked at whether species were nearer the “optimum” trait expression in highly suitable habitats (captured with community weighted means). We used cluster analysis on 138 plots located along a wet-to-dry ecotone in the Sandhills region of North Carolina based on several environmental factors including soil, nutrient and disturbance characteristics. Three environmental clusters were identified. We used indicator species analysis (ISA) to identify indicator species for each environmental cluster. Twelve species were selected, and for each species, plots were grouped into either “indicator” or “non-indicator” groups based on the results of ISA. Using two traits that we expected to be shaped by the environment (SLA and LDMC), we asked three questions: 1) Do species exhibit different average trait values within indicator and non-indicator plots? 2) Do species exhibit differing amounts of variation in indicator vs. non-indicator plots? 3) Are species’ trait means nearer to “optimum” trait values when they are in indicator plots?

Results/Conclusions

We found that 1/3 of species had significantly different mean trait values between the two groups. Only two species exhibited significantly different variation for LDMC between the two groups, while four species differed for SLA. Four species had traits that were closer to CWMs in the indicator group for LDMC, while three species had average trait values that were closer for SLA. One species showed average trait values that were significantly further from CWM in the indicator group for both traits. Our results indicate that species can demonstrate differing trait expression in habitats where they are widespread and more abundant versus habitats where they are present, but less abundant. However, the lack of consistently directional response indicates that expression of traits is shaped by both abiotic and biotic filters simultaneously, altering the direction of trait response to the environment. These findings provide further evidence that intraspecific variation should be considered in studies evaluating competition, coexistence and community assembly, and should be included in studies that attempt to understand and quantify ecosystem function.