Pattern and process at a desert grassland-shrubland ecotone

Abstract

Over the past century, shrub encroachment has occurred in desert grasslands all over the American Southwest. In the Northern Chihuahuan desert, most encroachment was due to creosotebush (Larrea tridentata [Sessé & Moc. ex DC.] Coville) expanding into black grama (Bouteloua eriopoda [Torr.] Torr.) dominated grasslands. Dominant plants contribute significantly to ecosystem structure and function, and grasses and shrubs differ fundamentally in their role in ecosystems. This dissertation investigated the influence of dominant plants, grasses and shrubs, on pattern and process at a desert grassland-shrubland ecotone. First, small-scale species coexistence patterns in microsites around dominant grasses (B. eriopoda) and shrubs (L. tridentata) were compared. Second, small-scale patterns were related to landscape scale species richness. Third, a gap-dynamics simulation model (ECOTONE) was used to investigate mechanisms that could create some of the patterns observed. Contrary to expectations, subdominant species abundance was lower in microsites around grasses than in microsites around shrubs. Different functional groups of subdominant species had a different distribution in microsites. When extrapolating to the landscape, it was found that perennial species richness is related to the distribution of dominant plants across scales from the plant to the landscape, whereas annual species richness seems to be mainly influenced by disturbance. Both biotic and abiotic factors contributed to the patterns observed across scales. Soil texture differences between microsites may explain differences in subdominant species abundance between the canopy area of plants and interspaces. In contrast, differences in subdominant species abundance between canopy areas of grasses and shrubs were most likely due to biotic factors, namely the morphological differences between the canopies of the dominant lifeforms. The use of the simulation model showed that black grama dominates the vegetation by a combination of intensive exploitation of water resources underneath its canopy and by its capacity to cope with soil disturbances, whereas creosotebush dominates by its long lifespan and low yearly biomass turnover rate. Replacement of black grama by creosotebush under the current disturbance and climatic regime would take a very long time. Similar to field observations, subdominant species abundance was higher in simulated creosotebush dominated landscapes than black grama dominated ones. It was concluded that the shift in the dominant lifeform by itself does not have a negative influence on plant species diversity at the ecotone investigated. On the contrary, the addition of shrubs to the vegetation increases spatial heterogeneity and has a positive effect on the overall diversity of the vegetation in an ecotone situation. Indirect processes induced by creosotebush that likely can be linked to species loss seem to only be important in areas where creosotebush dominates large extents of vegetation. Furthermore, other processes than shrub encroachment, such as human activities and drought, may contribute to species loss during desertification.