Predicting plant compositional responses to grazing with a functional traits model

Background/Question/Methods

Plant functional classification schemes arrange plants according to their patterns of morphological, physiological, and phenological traits and they attempt to establish response functional groups by aggregating those species with the same behavior in response to environmental disturbances such as grazing. Plant traits have been variously correlated with grazing along broad environmental gradients during the past 2 decades, but the prediction of species compositional change to large herbivores grazing in specific communities has received far less attention. We have developed a predictive framework of plant response to grazing that is founded upon the a priori approach to functional group construction that can be applied to diverse plant communities to inform management of grazed ecosystems. This framework is interfaced with the PHYGROW plant simulation model which provides the capacity to incorporate soil and climatic conditions into plant responses and to predict temporal patterns of community composition change at various grazing intensities.

Results/Conclusions

This trait-based framework integrates various ecological approaches and concepts across increasing ecological scales because: 1) it is based on homogenous ecological sites to minimize confounding effects among trait responses due to environmental heterogeneity, 2) it translates the avoidance and tolerance strategies of grazing resistance into trait categories based on empirical data and ecological theory, 3) it ranks the relative survival and performance of plants in response to grazing according to their possession of unique categories of tolerance and avoidance traits, 4) it scales the concept of grazing resistance from the traits associated with individual plants to entire communities, and 5) it attempts to link grazing response traits with plant strategies and plant performance at various grazing intensities. Comparing grazing response groups as outcomes of model simulations with field data we show various options of model adjustment to specific environments and grazing intensities.