| Abstract: | We describe a case study evaluating the ecological impact of Bromus tectorum L. (cheatgrass) invasion following fire disturbance and the effectiveness of revegetation as a means of rehabilitation in a degraded semiarid shrubsteppe system. The effectiveness of rehabilitation efforts was assessed relative to arthropod richness, vegetation and arthropod community composition, and ground-cover characteristics in 3 habitats: undisturbed, burned and weed infested ( B. tectorum ), and burned then rehabilitated with native and nonnative vegetation. Arthropods were collected in each habitat using pitfall traps. Differences in arthropod richness were compared using rarefaction curves. Nonmetric multidimensional scaling and nonparametric multivariate statistical procedures, including analysis of similarity and similarity percentage routines, were used to compare arthropod and vegetation community composition and ground-cover characteristics between habitats. Arthropod communities in the rehabilitated habitat were distinct from those observed in the undisturbed and weed-infested habitats. Rehabilitation in this study resulted in a shift toward conditions observed in an undisturbed habitat and perhaps is an intermediate step to complete restoration. Arthropod richness, arthropod and vegetation community composition, and ground-cover characteristics were all useful indicators but returned slightly different results. Assessing multiple variables yielded the most complete understanding of the habitats studied. |
