Using species functional traits to predict community dynamics
Date
2012
Authors
Ames, Gregory Michael, author
Webb, Colleen, advisor
Poff, N. LeRoy, committee member
Knapp, Alan, committee member
Noon, Barry, committee member
Journal Title
Journal ISSN
Volume Title
Abstract
A major goal for community ecology has been to determine a general set of rules to explain the structure and function of communities. Traits-based methods for describing community dynamics have been touted as providing a set of general methods to describe the structure and function of communities based on measurable properties of individual organisms in the community in a changing environment. Validation of traits-based methods that describe changes in community structure as a function of the interaction between functional traits along changing environmental gradients in real systems is needed. Here we present studies of three different plant communities where we use novel applications of traits-based Bayesian hierarchical models and principal component analysis to explain the changes in community structure/function and demonstrate that the communities are primarily structured by traits and their interactions with a changing environment. In a natural tallgrass prairie we were able to explain more than 84% of the variation in community functional diversity and an average of 64% of the cover variation across the ten species in the study over a 25-year span (Chapter 1). Additionally we show that changes in community structure are primarily explained by relative growth rate and its interaction with precipitation. In an experimentally manipulated grassland, our model explains more than 75% of the variation in total plot biomass over the course of 18 years. Further, we found that this system was primarily driven by the same trait/environment interactions as the tallgrass system. Finally, we show that trait/environment interactions allow us to explain 91% of the variation in plot biomass in a restored riparian wetland. Our ability to explain large portions of the variation in community structure and performance of these three distinct types of plant communities, using similar traits and environmental drivers, provides evidence of general laws underlying the structure of plant communities. This work represents a significant step toward understanding those general laws and helping community ecology develop from a largely descriptive science to a predictive science.
Description
Rights Access
Subject
modelling
Konza
traits
Bayesian
Cedar Creek