Effects Of Plant Community Structure, Soil Components, And Aboveground Fungal Presence On Grassland Productivity

This dissertation investigates how aspects of the resident plant community affect grassland productivity. Chapter One gives an overview of how grassland productivity can be affected by the structural components of a plant community, abiotic and biotic soil components, and the presence of aboveground fungi. In Chapter Two, I present results of an experiment where the frequency of interspecific interactions in plant communities was altered along richness and evenness gradients by either randomly placing species in plots (dispersed plots) or by aggregating species in groups of four individuals (aggregated plots). Results suggest aggregation decreased productivity by promoting species coexistence and not by decreasing niche partitioning and facilitation. In Chapter Three, I compare two diversity effect modeling approaches (additive partitioning model and Diversity Interaction models) and show how using sown and realized proportions may alter outcomes and interpretations of diversity effects analyses. In Chapter Four, I describe a set of experiments to determine whether soil feedbacks affect grassland species monoculture yields. To determine the mechanism (abiotic or biotic), focal species were grown in soil cores from conspecifically conditioned plots that removed soil biota by two different heating treatments. Results reinforce the facilitative effect of legumes and suggest nutrient limitation may be more important than soil biota effects in the early years of grassland establishment. In Chapter Five, I evaluated the effects of aboveground fungal presence in Pascopyrum smithii (western wheatgrass) and one of its’ cultivars, Rodan wheatgrass. Results suggest fungal presence affects multiple above- and belowground responses. However, the lack of specificity of the fungal presence testing method created difficulty in interpreting the results. I recommend the use of multiple methods to determine specific fungal presence as to ensure the identity of treatments being applied in experiments