Does an exotic invasive grass facilitate the invasion of a woody species into remnant prairies? A study of the native, Prosopis glandulosa and the alien, Sorghum halepense.

摘要

The native tree, Prosopis glandulosa (honey mesquite), and the exotic grass, Sorghum halepense (Johnsongrass), have been successfully invading native prairies in the Southwest U.S. since the 1800s. Although both species employ complementary life history traits in order to establish and flourish in grasslands, a correlation between the two has yet to be established. MacDougall and Turkington's (2005) "driver" or "passenger" model of invasive species was applied to explore mechanisms of invasion and impacts on community structure of these two species. My first objective was to determine if S. halepense was driving shifts in the abiotic and biotic structure of a native prairie community. Assuming that S. halepense was in fact altering both the resources and the plant neighborhood of the community, my second objective was to determine if S. halepense was facilitating the invasion of P. glandulosa . I predicted that in an invaded prairie zone, S. halepense (A) increased light levels, (B) decreased soil nitrogen (N) and increased soil carbon (C), and (C) decreased soil moisture relative to conditions in a prairie without this species, and thus (D)created a less diverse plant community, comprised of different plant species, and (E) facilitated the invasion of P. glandulosa, a known N-fixer. In 2005 a field experiment was established in a remnant tallgrass prairie in North Texas comprised of three distinct zones: a native zone dominated by the native grass S. scoparium, an invaded zone of S. halepense, and a mixed zone where both native and exotic grasses were present. A greenhouse competition experiment followed in 2006 comparing P. glandulosa's germination and growth in two soil types, native and invaded, with three levels of competition: no competition, S. scoparium, and S. halepense. To test predictions regarding the first objective, light levels, soil N and soil C were examined in both the field and the greenhouse while soil gravimetric moisture and plant community composition were measured in the field only. The presence of S. halepense significantly lowered light levels and increased soil N and soil moisture in the field plots relative to native prairie. However, the lowest light levels in the greenhouse experiment occurred in pots with S. scoparium owing to the presence of standing dead. Somewhat surprisingly, plant species diversity and the number of woody species were higher in the invaded zone than the native zone.;To address the second objective, P. glandulosa germination and seedling growth were measured in both the field and greenhouse. Ambient litter inhibited P. glandulosa germination in the field regardless of neighbors, although a correlation with decreased light was not established. Significantly more seeds germinated in the invaded zone then either the mixed or native zones in the field, while seeds planted in greenhouse pots with S. halepense germinated better then those planted alone or with S. scoparium, regardless of soil type. Contrary to the field germination results, there was no effect of litter on seedling growth, while in both the field and greenhouse competition from live neighbors inhibited seedling growth. In the field seedlings in the native zone without competition grew the tallest, and seedling mortality caused by herbivores was significantly higher in the invaded zone (60-80%) than the native zone (20%). In the greenhouse seedling growth was equally inhibited by the presence of S. halepense and S. scoparium.;These results suggest that the soil abiotic resources and community structure were being altered in a way that facilitated seedling recruitment but not growth of P. glandulosa, and the shift from grassland to woodland was multivariate in origin. Anthropogenic influences, such as disturbance, fragmentation, and land management, may have caused the low diversity in the native zone and may have interacted with S. halepense's indirect effects on soil moisture, light, and N availability to drive community level changes in this remnant prairie.