Scaling-up trait covariation: Coordination and trade-offs within and among plant species in alpine meadow communities

Abstract

<p indent="0mm">Functional traits and their covariation could reflect specific ecological strategies of organisms in response and adaptation to changing environment as the basis of understanding organism-environment relations at the individual, population, and community levels. Many studies report trait coordination and trade-offs among species over broad scales, especially for the leaf economic spectrum at regional and global scales. An increasing number of studies have attempted to explain and predict the functional (trait) structure of populations, communities, and vegetation in response to environmental change, but the mechanisms underpinning trait coordination and trade-offs may differ across ecological levels. We know little about this trait coordination and trade-offs within and among species in local communities, lacking not only relevant theoretical frameworks but also experimental evidence. Merging trait covariation into theory for community assembly, we predict: (1) Significant trait coordination or trade-offs among coexisting species occur when coexisting species occupy distinct ecological niches, similar to the pattern induced by specialization of species niches on a broad scale and (2) if trait covariations are induced by functional balance and/or structural optima, then consistent trait coordination and trade-offs within and among species will occur and scale up to the community and ecosystem levels. In this study, we experimentally test these predictions in local communities and discuss the potential mechanisms underpinning trait coordination and trade-offs from the scale of individual plants to vegetation. In Tibetan alpine meadows with high plant species richness but low environmental heterogeneity, we sampled two leaf stoichiometric traits (leaf nitrogen and phosphorus content), two leaf morphological traits (specific leaf area and leaf dry matter content) and mature plant height for three to five individuals for all coexisting species within each of 36 plant communities (plots) distributed along a gradient of environmental harshness. In each plot, we averaged trait values of individual samples of each species at the population-level, and used the mean of all populations of each species to indicate the species-level traits. We assessed pairwise trait correlation within and among 29 frequent species by standardized major axis regression and trait coordination and trade-offs among multiple traits by principal component analysis. Similar to the broad-scale pattern, we found that leaf nitrogen and phosphorus contents were significantly positively correlated with specific leaf area (negatively correlated with leaf dry matter content) among 29 coexisting species, but these trade-offs were significant at a population level in only a few species. Both among species and within most species, (1) leaf nitrogen content was significantly positively correlated with phosphorus content, likely induced by functional balance, (2) specific leaf area was negatively correlated with leaf dry matter content, indicating a structural optimum, (3) neither leaf stoichiometric nor morphological traits significantly correlated with plant height. Trait coordination and trade-offs were stronger for multiple trait relationships than for bivariate correlations. These results support our predictions and provide a basis and experimental evidence for scaling up trait covariations from individuals to communities.