In the southern Great Plains, the encroachment of grassland ecosystems by mesquite (Prosopis glandulosa), is widespread, and prescribed fire is commonly used in its control. Despite this, substantial quantitative information concerning their influences on the community composition, functional dynamics, and soil organic carbon (SOC) storage potential of grassland ecosystems is lacking. The objectives of this study were to: a) quantify the effects of seasonal prescribed fire treatments and mesquite encroachment on aboveground net primary productivity (ANPP) and herbaceous community composition; b) characterize SOC pool sizes, turnover, and storage potential relative to vegetation type and fire treatment; c) evaluate the structure and diversity of soil microbial communities relative to vegetation type; and d) characterize the functional diversity of these same microbes using the GeoChip functional gene microarray. Repeated winter and summer fires led to increased ANPP rates (average, 434 and 313 g m-2 y-1, respectively), relative to unburned controls (average, 238 g m-2 y-1), altered herbaceous community composition, and increased the storage of resistant forms of SOC, but did not affect overall SOC storage. Herbaceous ANPP rates did not differ significantly as a result of mesquite encroachment, but herbaceous community composition and SOC storage did. Mesquite soils contained significantly more total, slow-turnover, and resistant forms of SOC than those that occurred beneath C3 or C4 grasses. Similarity among the soil bacterial and fungal communities associated with the major vegetation types in this system was low to moderate. Significant differences were detected among soil fungi, with the mesquite-associated fungi harboring significant differences in community structure relative to the fungal communities associated with each of the other vegetation types examined. Despite this result, few significant differences were detected with respect to the functional diversity of these communities, suggesting either a high degree of functional redundancy, or that the functional differences harbored by these communities are beyond the scope of the GeoChip. The results of this study demonstrate that both fire and mesquite encroachment have the potential to alter ecosystem components and processes significantly, providing new insight regarding the effects of these widespread land use and land cover changes on ecosystem structure and function.

<p>Since the pre-industrial age, the Earth has been warming at unparalleled rates, and this warming is changing climate and weather, creating a more extreme global hydrological cycle. In this dissertation, I explore how these changes to the hydrological cycle may act ecosystem and community level responses of terrestrial plants in the Midwestern United States. In this region, it is projected that mean annual precipitation (MAP) will increase, but precipitation will become more variable across and within seasons. Ecosystem structure and function are vulnerable to changes in hydrologic patterns, including changes in biogeochemical cycles, plant productivity, and plant community structure and function. In this dissertation, I explore how changes in precipitation will alter these processes using two field experiments, and I suggest potential hypotheses that could explain drought-induced community change.</p><p><br></p><p>In chapter 1, I explore how alterations to seasonal precipitation in the winter and summer act ecosystem and community processes in a temperate deciduous forest. Biogeochemical processes and plant communities are sensitive to changes in abiotic conditions, and these conditions will alter forest succession, particularly juvenile woody plant species. Using a fully factorial experiment, I manipulated winter snowfall and summer precipitation to create wet, dry, and control (ambient conditions) treatments and investigated how changes in seasonal precipitation would act mineralization rates, woody plant recruitment, and understory composition. I found that the effects of winter and summer precipitation on these processes acted independently of one another in this system, and the system was resistant to changes in mineralization rates and understory composition. Woody plant recruitment may be more sensitive to altered precipitation, as recruitment of at least one of the four species planted, Lindera benzoin, was impacted by changes in seasonal precipitation. Snow removal treatments reduced germination and increased summer precipitation decreased the relative growth rate of this species. In the short term, slight changes to woody plant recruitment may have little impact on long-term forest succession, but as these changes persist over longer periods of time, they could alter the direction of succession, which could lead to changes in the understory community composition and nutrient cycling.</p> <p><br></p><p>The second and third chapters explore the effects that drought intensification will have on terrestrial plant communities. Numerous studies have investigated the effects of individual droughts on ecosystem and community responses, but the effects that both the timing and duration of drought have on these responses remain largely unknown. To explore this gap in the literature, I conducted a eld experiment using rainout shelters to reduce growing season precipitation, creating dry periods that varied in length and timing. Drought can impact productivity and diversity in this system, and the timing in which the drought occurs influences these effects. Surprisingly, I found that the length of drought did not affect productivity or community composition.</p> <p><br></p><p>The final chapter introduces the Community Response to Extreme Drought framework CRED), which addresses the potential temporal progression of mechanisms and plant-plant interactions that may lead to community changes during and after a drought. The mechanisms for the temporal evolution of community-level drought responses are not fully understood, but plant-plant interactions, both competitive (-) and facilitative (+), are increasingly being recognized as important drivers of community compositional changes. The CRED framework provides hypotheses for the roles that plant-plant interactions have on drought-induced community change. CRED addresses how system-specific variables and the intensity of drought may influence the strength of plant-plant interactions over time, and ultimately the systems resistance and resilience to drought. </p><p><br></p><p>The results from this dissertation work have revealed that more research needs to be done to fully understand how changes in precipitation regimes and patterns will affect terrestrial ecosystems and plant communities. A better understanding of how ecosystems and communities respond to drought timing and length can help improve climate models and restoration strategies.</p>