|Abstract or Summary
- Biological plant invasions are diminishing the ecological integrity and function of ecosystems worldwide. A primary example of this is in the Great Basin of the United States, where invasive annual grasses, like cheatgrass (Bromus tectorum L.) and medusahead (Taeniatherum caput-medusae L. Nevski), are dominating many sagebrush-steppe ecosystems. In these invaded areas, restoration is fundamental to recovering plant community structure and function of once productive and diverse systems. Seeding is a management practice that accelerates revegetation of perennial plants in annual grass invaded ecosystems. However, seeding generally produces poor establishment and patchy plant communities that are quickly replaced by invaders. Identifying and overcoming the processes limiting desirable plant recovery should increase restoration success in annual grass invaded ecosystems. Seed dispersal dynamics, including propagule pressure and priority effects, strongly influence plant establishment and community assembly, especially during invasion and restoration because they can control recruitment of newly arriving species by influencing safe site occupation. The overall objective of this research was to identify the role of propagule pressure and priority effects on seedling emergence, plant life history, and plant community assembly.
My study involved three experiments that were evaluated on seeded annual grass invaded sagebrush-steppe ecosystems in eastern Oregon. First, in order to identify the role of propagule pressure, or number of viable seeds in the seed bank, and annual grass dispersal timing, I conducted a factorially arranged experiment that tested the priority effects of seeding annual grasses in autumn or spring, adding water, and varying annual and perennial grass propagule pressure by seeding 150, 1,500, 2,500, or 3,500 seeds m⁻². My results suggested that providing perennial grasses a priority seeding effect by delaying annual grass seeding until spring initially facilitated perennial grass density; however, this effect did not persist into the second-year following seeding. In addition, when annual grass propagule pressure exceeded 150 seeds m⁻², an ecological threshold was crossed which limited perennial grass recruitment regardless of perennial grass seeding rate. When water availability was high, perennial grass establishment was high because safe site availability increased, but perennial grass establishment depended on annual grass propagule pressure. These results demonstrated that restoring perennial grasses to annual grass invaded ecosystems may be possible when perennial grass seeding rates and water availability are high. However, if annual grass propagule pressure exceeds 150 seeds m⁻², an ecological threshold occurs, where, perennial grass recruitment will be limited regardless of seeding strategy.
A second factorial experiment was designed to identify the priority effects of perennial grass seeding time and frequency, annual and perennial grass propagule pressure, and water availability on seedling emergence and annual and perennial grass density and biomass two-years following seeding. In this experiment, my results indicated that seeding perennial grasses seasonally split between the autumn and spring
produced higher perennial grass density and biomass compared to seeding perennial grasses exclusively in either period. In addition, results supported my hypothesis that perennial grass density and biomass in seasonally split applications would be higher where perennial propagule pressure was high and annual grass propagule pressure was low. However, I found that there was a threshold between 150-1,500 annual grass seeds m⁻², where regardless of perennial grass seeding strategies, perennial grass density and biomass was low. When water was added, annual and perennial grass density was higher than in plots without additional water, suggesting that higher water availability facilitates the growth of all seeded species. Collectively, these results suggested that modifying perennial grass seeding times and frequency increased perennial grass recruitment to annual grass invaded ecosystems, but only if annual grass propagule pressure was below 1,500 annual grass seeds m⁻².
Third, a life history approach was used to identify and quantify the effect of ecological processes on plant population demography when annual grass seeding times varied (autumn or spring), annual and perennial grass propagule pressure was modified by 150, 1,500, 2,500, or 3,500 seeds m⁻², and watering occurred (ambient or water added treatments). In this study, we found that all species had low emergence rates, even though seedling germination was relatively high. Based on prior research, this suggests that freeze-thaw cycles, pathogen attack, and soil crusts may strongly inhibit plant growth from the germination to emergence growth stages. Alternatively, my finding that perennial grass germination rates were higher when they were seeded with annual grasses in autumn compared to delaying annual grass seeding until spring suggests that perennial grass germination is facilitated by neighboring annual grasses during this life history
stage. Following seedling emergence, my data indicated that adding water enhances the establishment and growth of all species and that providing perennial grasses a priority seeding effect by delaying annual grass seeding until spring yields higher perennial grass density. However, delaying annual grass seeding until spring only provided perennial grasses a priority effect when annual grass propagule pressure was high. When annual grass propagule pressure was low, seeding perennial grasses in autumn yielded the highest perennial grass density through their life history suggesting low numbers of neighboring annual grasses facilitates the density of perennial grasses. By the second growing season, annual grass density was two-times higher than initial annual grass seeding rates and over four-times higher than perennial grass density, suggesting that annual grass interference may increase during the second growing season. Plant community assembly of restored shrub-steppe ecosystems degraded by annual grasses will likely be determined by the establishment of seeded species in the first growing season.
Because perennial grasses are facilitated by annual grasses in the germination stage, and by small numbers of neighboring annual grasses in later growth stages, if perennial grasses establish in areas with low annual grass propagule pressure in the first growing season, they will likely persist to become adults. Alternatively, high annual grass propagule pressure limits perennial grass recruitment regardless of seeding strategy because of their preemptive occupation of safe sites and soil resources and high density by the second growing season.