The impact of shifting precipitation patterns on the establishment and biomass allocation patterns of shrub and grass seedlings in southwest Texas

Abstract

Anthropogenic emissions contribute to an annual 0.5% increase in atmospheric CO₂. As global CO₂ levels increase, regional precipitation patterns in southern Texas will likely be altered, resulting in an increase in winter/spring precipitation and a decrease in summer precipitation. My primary objective was to determine whether this change in precipitation seasonality would confer a competitive advantage to the establishment of either grasses or shrubs, and how grazing intensity and light availability might interact with precipitation patterns to influence grass and shrub seedling establishment and biomass allocation. Seedlings of 3 shrub species (Prosopis glandulosa var. glandulosa, Acacia berlandieri, and A. wrightii var. greggii) and 3 grass species (Aristida purpurea var. wrightii, Stipa leucotricha, and Setaria texana) were subjected to water regimes based on probable precipitation patterns in a CO₂ enriched atmosphere (0.6, 0.8, and 1.0 ambient precipitation in summer and 1.0, 1.15, and 1.3 ambient precipitation in winter). Seedlings were grown under two levels of light availability (100 vs. 50% ambient) and subjected to three defoliation treatments (non-defoliated, light defoliation, and heavy defoliation). Plants were harvested at regular intervals and the leaves, stems, and roots were dried and weighed separately. Reducing summer precipitation and increasing winter/spring precipitation generally did not significantly affect the growth or biomass allocation patterns of grass and shrub seedlings. Shrub seedlings exhibited greater biomass allocation plasticity during the summer, and grass seedlings exhibited greater plasticity during the winter/spring. Defoliation intensity and microsite reductions in light availability influenced seedling establishment and biomass allocation patterns more than reducing summer or enhancing winter/spring precipitation. My second objective was to determine whether hydraulic lift occurs in mesquite. Six mature mesquite trees were randomly assigned to 3 treatments. Two mesquite were cut down, removed, and shade cloth was used to mimic canopy effects. Two mesquite received subsurface irrigation, and two mesquite served as controls. Diel fluctuations in soil water potential characteristic of hydraulic lift were observed, but I was unable to conclusively isolate hydraulic lift by mesquite as the source of these fluctuations in soil water potential.